© 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. 776
    Prev
    Next
    INDEX
    Explanations

    hints or subtle indications

    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
    99
    +0.11
    0.5%
    506
    +0.10
    0.5%
    129
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    129
    +0.11
    0.03
    413
    +0.10
    0.03
    195
    +0.10
    0.02
    Negative Logits
    rament
    -0.76
    ategory
    -0.68
    scrib
    -0.68
    quished
    -0.66
    ourses
    -0.66
    miah
    -0.65
    portion
    -0.65
    die
    -0.65
    raid
    -0.63
    queue
    -0.63
    POSITIVE LOGITS
     hint
    1.47
     hints
    1.33
     glimps
    1.11
     clue
    1.03
     clues
    1.00
     hinted
    0.92
     glimpse
    0.87
     signs
    0.83
     intim
    0.80
     indications
    0.77
    Activations Density 0.187%

    No Known Activations