© 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. 10168
    Prev
    Next
    INDEX
    Explanations

    proper names like "Pat", especially at high activations

    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
    218
    +0.11
    0.5%
    283
    +0.11
    0.5%
    81
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    81
    +0.11
    0.02
    589
    +0.11
    0.02
    461
    +0.10
    0.02
    Negative Logits
    vous
    -0.79
     interception
    -0.68
     Monstrous
    -0.67
    ADE
    -0.67
    xual
    -0.65
    hower
    -0.65
    STD
    -0.63
     Scouting
    -0.62
    Analysis
    -0.61
     equation
    -0.61
    POSITIVE LOGITS
    ricia
    1.42
    riots
    1.31
    riot
    1.31
    rice
    1.29
    hetic
    1.24
    terson
    1.23
    rolled
    1.22
    hetically
    1.18
    reon
    1.17
    ric
    1.14
    Activations Density 0.040%

    No Known Activations