© 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 768
    9. 8-RES_FS768-JB
    11. 732
    Prev
    Next
    INDEX
    Explanations

    details related to historic events and people mentioned in news articles

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    27
    +0.12
    0.5%
    765
    +0.11
    0.5%
    171
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    27
    +0.12
    0.11
    362
    +0.11
    0.15
    36
    +0.11
    0.17
    Negative Logits
    blems
    -0.79
    ropolis
    -0.69
    omination
    -0.69
    licts
    -0.68
    achelor
    -0.67
    inctions
    -0.67
     :=
    -0.64
    bernatorial
    -0.64
    ignty
    -0.64
    urden
    -0.63
    POSITIVE LOGITS
     standpoint
    1.02
     vantage
    0.87
     ausp
    0.84
     intermediary
    0.82
     margin
    0.76
     outset
    0.72
     margins
    0.69
     supplied
    0.69
     perspective
    0.67
     persuasion
    0.67
    Activations Density 8.251%

    No Known Activations