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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 25669
    Prev
    Next
    INDEX
    Explanations

    repeated instances of the word "continue."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    515
    +0.11
    0.5%
    201
    +0.11
    0.5%
    62
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    62
    +0.11
    0.02
    79
    +0.11
    0.02
    355
    +0.11
    0.02
    Negative Logits
    istani
    -0.78
    nikov
    -0.75
    istically
    -0.72
    esp
    -0.72
    typ
    -0.71
    ariat
    -0.69
    onom
    -0.69
    otom
    -0.68
    bish
    -0.68
    utive
    -0.68
    POSITIVE LOGITS
     BELOW
    0.76
     Reading
    0.75
     Below
    0.75
     progressing
    0.74
     scrolling
    0.73
     reading
    0.68
     below
    0.67
     Albion
    0.64
     banging
    0.63
    Loading
    0.61
    Activations Density 0.016%

    No Known Activations