Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 1765
    Prev
    Next
    INDEX
    Explanations

    descriptions of physical confrontations or aggressive actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    descriptions of group violence or aggressive actions.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @trainjumperx

    violent physical altercations or assaults.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @trainjumperx

    descriptions of group violence or aggression where people attack, beat, or threaten someone.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @trainjumperx
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    198
    +0.10
    0.3%
    1531
    +0.09
    0.3%
    946
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    946
    +0.10
    0.06
    222
    +0.09
    0.06
    1490
    +0.09
    0.06
    Negative Logits
     sement
    -0.75
     pól
    -0.74
     makro
    -0.73
     broder
    -0.71
     sevilla
    -0.71
     anse
    -0.71
     santiago
    -0.70
     conector
    -0.67
     siena
    -0.67
     lavander
    -0.67
    POSITIVE LOGITS
     themselves
    0.85
    themselves
    0.82
     their
    0.70
    their
    0.66
     collectively
    0.64
     THEIR
    0.63
    Their
    0.59
     leurs
    0.55
     philosophic
    0.54
     Their
    0.54
    Activations Density 0.572%

    No Known Activations