© 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 12288
    9. 8-RES_FS12288-JB
    11. 1172
    Prev
    Next
    INDEX
    Explanations

    phrases related to people coming together in groups or crowds

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    756
    +0.11
    0.5%
    677
    +0.11
    0.5%
    216
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    677
    +0.11
    0.03
    271
    +0.11
    0.03
    216
    +0.10
    0.03
    Negative Logits
    void
    -0.71
    price
    -0.69
    ble
    -0.68
    ysis
    -0.66
    codes
    -0.66
    plays
    -0.65
    code
    -0.64
    word
    -0.63
    bler
    -0.63
    hole
    -0.63
    POSITIVE LOGITS
     together
    0.87
     gathered
    0.85
     signatures
    0.84
    ivities
    0.81
    ãĤ¤ãĥĪ
    0.74
     gathers
    0.73
    roup
    0.73
     gather
    0.72
    roups
    0.72
     gathering
    0.71
    Activations Density 0.098%

    No Known Activations