© 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 24576
    9. 8-RES_FS24576-JB
    11. 23856
    Prev
    Next
    INDEX
    Explanations

    nouns related to ranking or listing accomplishments or statistics

    oai_token-act-pair · gpt-4o-miniTriggered 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
    578
    +0.14
    0.7%
    679
    +0.12
    0.5%
    123
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.14
    0.03
    679
    +0.12
    0.05
    578
    +0.10
    0.03
    Negative Logits
    sw
    -0.78
    aly
    -0.76
    Otherwise
    -0.70
    SHIP
    -0.68
    Õ
    -0.65
    thumbnails
    -0.64
    EMBER
    -0.64
    اÙĦ
    -0.64
    alist
    -0.63
    usercontent
    -0.62
    POSITIVE LOGITS
    omething
    1.18
    ometimes
    1.18
    pace
    1.15
    hift
    1.14
    creen
    1.12
    ilver
    1.11
    heet
    1.09
    hare
    1.03
    paces
    1.01
    uggest
    1.00
    Activations Density 0.269%

    No Known Activations