© 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 · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 1772
    Prev
    Next
    INDEX
    Explanations

    verbs relating to actions or decisions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    218
    +0.11
    0.5%
    665
    +0.10
    0.5%
    161
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.08
    608
    +0.10
    0.08
    123
    +0.10
    0.06
    Negative Logits
    Released
    -0.68
    terday
    -0.66
    uably
    -0.63
    xtap
    -0.63
     debuted
    -0.62
    iom
    -0.61
    lda
    -0.60
    undrum
    -0.58
     premiered
    -0.57
    aird
    -0.57
    POSITIVE LOGITS
     oneself
    0.98
     anything
    0.98
     whatever
    0.97
     whichever
    0.96
     somebody
    0.88
     whoever
    0.86
     any
    0.84
     them
    0.84
     someone
    0.82
     anybody
    0.82
    Activations Density 0.508%

    No Known Activations