© 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 3072
    9. 8-RES_FS3072-JB
    11. 2994
    Prev
    Next
    INDEX
    Explanations

    verbs related to motivation and inspiration

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    69
    +0.11
    0.5%
    255
    +0.11
    0.5%
    149
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.11
    0.08
    271
    +0.11
    0.05
    69
    +0.10
    0.06
    Negative Logits
    çĦ
    -0.70
    bill
    -0.65
    rock
    -0.64
    psons
    -0.64
    arant
    -0.63
    arent
    -0.62
     ç¥ŀ
    -0.62
     overrun
    -0.60
    oland
    -0.60
    puted
    -0.60
    POSITIVE LOGITS
     curiosity
    0.90
    ingly
    0.82
     impulses
    0.76
     youngsters
    0.76
     imagination
    0.75
     impulse
    0.72
     laughter
    0.72
     creativity
    0.72
    Reviewer
    0.71
     Gree
    0.71
    Activations Density 1.122%

    No Known Activations