Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1701
    Prev
    Next
    INDEX
    Explanations

    technical or software-related terms and browser-related information

    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
    224
    +0.11
    0.5%
    484
    +0.10
    0.5%
    316
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.15
    326
    +0.10
    0.12
    659
    +0.10
    0.09
    Negative Logits
    algia
    -0.97
    spring
    -0.82
    rums
    -0.81
    atics
    -0.78
    ere
    -0.77
    marks
    -0.76
    rum
    -0.74
    fare
    -0.73
    places
    -0.73
    alg
    -0.72
    POSITIVE LOGITS
     Hornets
    0.69
    sey
    0.68
     alley
    0.67
    INO
    0.66
    vernment
    0.65
     Hed
    0.64
    ï¸
    0.64
    ļéĨĴ
    0.64
     handc
    0.63
    vous
    0.61
    Activations Density 1.895%

    No Known Activations