© 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. 7227
    Prev
    Next
    INDEX
    Explanations

    financial and resource-related terms and numbers

    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
    447
    +0.19
    0.9%
    51
    +0.14
    0.6%
    724
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    724
    +0.19
    0.06
    51
    +0.14
    0.05
    255
    +0.13
    0.04
    Negative Logits
    CHAT
    -0.74
    umbn
    -0.68
    eatures
    -0.67
    cknow
    -0.61
    odcast
    -0.60
     Seas
    -0.60
     Massacre
    -0.57
     guiActiveUn
    -0.56
     IRC
    -0.56
    ART
    -0.56
    POSITIVE LOGITS
     than
    0.89
     apiece
    0.84
     allotted
    0.78
     per
    0.75
     upfront
    0.75
     compared
    0.74
     turnover
    0.74
     allocated
    0.74
     calories
    0.73
     overhead
    0.71
    Activations Density 0.178%

    No Known Activations