© 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. 317
    Prev
    Next
    INDEX
    Explanations

    phrases starting with 'You need' followed by a verb

    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
    8
    +0.10
    0.4%
    609
    +0.09
    0.4%
    52
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    8
    +0.10
    0.06
    609
    +0.09
    0.05
    52
    +0.09
    0.05
    Negative Logits
     Dism
    -0.62
    ihu
    -0.60
     Sod
    -0.56
     Zam
    -0.55
     Rost
    -0.55
     quir
    -0.54
    washer
    -0.54
     Democr
    -0.53
    anka
    -0.53
     Lama
    -0.53
    POSITIVE LOGITS
    lessly
    1.28
     patience
    0.85
    n
    0.85
     assurances
    0.76
     reminding
    0.75
     urgently
    0.75
    othes
    0.73
    nces
    0.71
     attention
    0.70
    FINE
    0.68
    Activations Density 0.104%

    No Known Activations