© 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. 3055
    Prev
    Next
    INDEX
    Explanations

    present-tense verbs

    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
    576
    +0.11
    0.5%
    404
    +0.11
    0.5%
    323
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.11
    -0.04
    232
    +0.11
    0.08
    284
    +0.10
    0.07
    Negative Logits
    auga
    -0.69
    bourg
    -0.63
     STATS
    -0.63
     confines
    -0.61
    utics
    -0.61
     goats
    -0.60
    ashtra
    -0.60
    »Ĵ
    -0.58
    htar
    -0.57
    vp
    -0.55
    POSITIVE LOGITS
    Wan
    0.87
    uay
    0.75
    worldly
    0.73
    uary
    0.72
    leck
    0.71
     Grip
    0.68
    Balt
    0.67
    wald
    0.66
    yssey
    0.64
    actory
    0.64
    Activations Density 1.117%

    No Known Activations