© 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. 10111
    Prev
    Next
    INDEX
    Explanations

    phrases indicating wrongdoing or negative actions

    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
    239
    +0.12
    0.5%
    13
    +0.11
    0.5%
    78
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    13
    +0.12
    0.05
    239
    +0.11
    0.03
    271
    +0.10
    0.03
    Negative Logits
    ipel
    -0.75
     Fever
    -0.70
     Arri
    -0.68
     Dwell
    -0.63
     tongues
    -0.62
     Torn
    -0.62
    aten
    -0.60
    Ent
    -0.60
    bris
    -0.60
     Mand
    -0.59
    POSITIVE LOGITS
     differently
    0.94
     wrong
    0.88
    wrong
    0.86
     offensively
    0.81
     administr
    0.81
     rash
    0.79
     proactive
    0.79
     harm
    0.78
     injustice
    0.76
     defensively
    0.75
    Activations Density 0.154%

    No Known Activations