© 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
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 1051
    Prev
    Next
    INDEX
    Explanations

    people's names, film titles, and creative roles like director or producer

    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
    1870
    +0.12
    0.3%
    1479
    +0.10
    0.3%
    50
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1479
    +0.12
    0.02
    429
    +0.10
    0.02
    1141
    +0.08
    0.02
    Negative Logits
     koncert
    -0.53
     sentito
    -0.53
     incroy
    -0.51
     klokken
    -0.49
     fondamentali
    -0.49
     lavorato
    -0.49
    ayuno
    -0.49
    MENAFN
    -0.48
     ModelExpression
    -0.48
     constate
    -0.48
    POSITIVE LOGITS
     director
    0.81
     directors
    0.78
     Directors
    0.71
    director
    0.71
     Director
    0.69
     filmmaker
    0.66
    directors
    0.66
     Directed
    0.66
     directed
    0.65
    Directors
    0.64
    Activations Density 0.100%

    No Known Activations