© 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. 7350
    Prev
    Next
    INDEX
    Explanations

    words related to physical or metaphorical building up

    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
    171
    +0.09
    0.4%
    207
    +0.09
    0.4%
    399
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.09
    0.03
    665
    +0.09
    0.03
    641
    +0.09
    0.03
    Negative Logits
    gee
    -0.81
    wered
    -0.79
    gnu
    -0.70
    oped
    -0.70
    write
    -0.70
    writers
    -0.69
    blers
    -0.69
    kov
    -0.68
    izon
    -0.67
    bler
    -0.65
    POSITIVE LOGITS
     momentum
    1.37
     rapport
    1.34
     tension
    1.11
     suspense
    1.10
     awareness
    1.10
     friendships
    1.06
     confidence
    1.05
     credibility
    1.01
     relationships
    1.01
     cohesion
    0.99
    Activations Density 0.100%

    No Known Activations