© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 2448
    Prev
    Next
    INDEX
    Explanations

    dates given in a news article

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    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
    1978
    +0.15
    0.5%
    674
    +0.13
    0.4%
    1343
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    315
    +0.15
    0.03
    189
    +0.13
    0.02
    45
    +0.12
    0.02
    Negative Logits
     intersper
    -1.55
     impra
    -1.51
     unspeak
    -1.46
     milf
    -1.42
     shenan
    -1.41
     increa
    -1.41
     hairc
    -1.41
     hentai
    -1.38
     gaily
    -1.34
     encomp
    -1.32
    POSITIVE LOGITS
     meras
    0.96
     optik
    0.95
     pól
    0.92
     alkoh
    0.89
     geograf
    0.89
     keramik
    0.89
     akut
    0.88
     solidar
    0.86
     silikon
    0.86
     sement
    0.85
    Activations Density 0.044%

    No Known Activations