© 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. 9275
    Prev
    Next
    INDEX
    Explanations

    personal thoughts and observations in a context involving named individuals

    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
    545
    +0.09
    0.4%
    374
    +0.09
    0.4%
    349
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    750
    +0.09
    0.03
    718
    +0.09
    0.04
    566
    +0.08
    0.03
    Negative Logits
     attest
    -0.66
     understandably
    -0.65
    eeper
    -0.63
     Nanto
    -0.62
     irrespective
    -0.60
     Attribution
    -0.60
     understandable
    -0.59
     reiterate
    -0.59
     ascertain
    -0.58
    catentry
    -0.58
    POSITIVE LOGITS
     existed
    0.81
     anymore
    0.77
     Clown
    0.67
     bothered
    0.64
     nor
    0.63
     happening
    0.63
     adolesc
    0.62
    !'"
    0.61
    ammed
    0.61
    ceivable
    0.61
    Activations Density 0.255%

    No Known Activations