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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 1496
    Prev
    Next
    INDEX
    Explanations

    legal and law enforcement-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_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
    373
    +0.12
    0.6%
    217
    +0.12
    0.6%
    498
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    217
    +0.12
    0.04
    498
    +0.12
    0.05
    720
    +0.10
    0.04
    Negative Logits
    olate
    -0.64
    oler
    -0.61
    cow
    -0.61
    olerance
    -0.60
    olated
    -0.60
    estate
    -0.59
    owed
    -0.58
     Wildcats
    -0.58
    crop
    -0.58
    UGE
    -0.57
    POSITIVE LOGITS
     unsurprisingly
    0.82
    among
    0.79
     interestingly
    0.76
     controvers
    0.74
     incidentally
    0.73
     alas
    0.73
     uh
    0.72
     moreover
    0.68
     notably
    0.68
     namely
    0.65
    Activations Density 0.117%

    No Known Activations