© 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 3072
    9. 8-RES_FS3072-JB
    11. 2979
    Prev
    Next
    INDEX
    Explanations

    phrases related to being caught or caught in a conflict or situation

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    1
    +0.15
    0.7%
    59
    +0.14
    0.7%
    314
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1
    +0.15
    0.06
    59
    +0.14
    0.05
    326
    +0.14
    0.05
    Negative Logits
    sburgh
    -0.64
    die
    -0.61
    arten
    -0.61
    bard
    -0.61
     pronoun
    -0.60
     princip
    -0.60
    inburgh
    -0.59
     sclerosis
    -0.59
    edly
    -0.58
    AMY
    -0.57
    POSITIVE LOGITS
    phrase
    1.06
    netflix
    0.82
    tails
    0.78
     glimps
    0.78
    izoph
    0.74
    doors
    0.73
     bait
    0.72
     Luffy
    0.71
    tail
    0.71
     unprepared
    0.70
    Activations Density 0.782%

    No Known Activations