© 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. 37306
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names and terms associated with notable individuals and places

    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
    481
    +0.18
    0.8%
    88
    +0.10
    0.5%
    231
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.18
    0.06
    659
    +0.10
    0.04
    588
    +0.10
    0.03
    Negative Logits
     IMP
    -0.77
     Force
    -0.76
     IB
    -0.76
     jails
    -0.76
     retarded
    -0.75
     FML
    -0.74
     DEF
    -0.74
     indo
    -0.72
     IM
    -0.69
     force
    -0.68
    POSITIVE LOGITS
    choes
    1.04
    atha
    0.99
    ét
    0.97
    rawl
    0.97
    ĸļ
    0.95
    onial
    0.93
    ón
    0.91
    ĸļ士
    0.91
    enei
    0.91
    yna
    0.89
    Activations Density 0.111%

    No Known Activations