© 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 49152
    9. 8-RES_FS49152-JB
    11. 26580
    Prev
    Next
    INDEX
    Explanations

    references to individuals or groups and their actions in a narrative context

    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
    656
    +0.11
    0.5%
    309
    +0.11
    0.5%
    169
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.11
    0.07
    354
    +0.11
    0.07
    762
    +0.10
    0.07
    Negative Logits
    undrum
    -0.78
     Consortium
    -0.69
     BI
    -0.68
    akeru
    -0.68
    EStream
    -0.67
     Eleven
    -0.66
    owder
    -0.65
    eligible
    -0.63
     Rebellion
    -0.63
    ibrary
    -0.61
    POSITIVE LOGITS
    'll
    0.85
    deals
    0.80
     strokes
    0.79
     conclud
    0.78
     caut
    0.77
    're
    0.75
     warn
    0.74
     encour
    0.74
    've
    0.74
     forbid
    0.72
    Activations Density 0.246%

    No Known Activations