© 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. 19928
    Prev
    Next
    INDEX
    Explanations

    legal or rule-breaking actions and consequences

    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
    326
    +0.12
    0.5%
    183
    +0.10
    0.5%
    549
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    220
    +0.12
    0.03
    260
    +0.10
    0.03
    310
    +0.10
    0.03
    Negative Logits
    olicy
    -0.86
     tremend
    -0.81
    atre
    -0.72
    ĺħ
    -0.71
    pmwiki
    -0.70
    enhagen
    -0.70
    pherd
    -0.69
    etsk
    -0.69
    rison
    -0.68
    onse
    -0.68
    POSITIVE LOGITS
    384
    1.30
    6666
    1.15
    th
    1.00
    05
    0.97
    07
    0.96
    09
    0.95
    03
    0.93
    06
    0.92
    08
    0.89
    02
    0.88
    Activations Density 0.037%

    No Known Activations