© 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. 7372
    Prev
    Next
    INDEX
    Explanations

    math-related terms and phrases

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    767
    +0.10
    0.4%
    254
    +0.09
    0.4%
    366
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    767
    +0.10
    0.03
    326
    +0.09
    0.04
    141
    +0.09
    0.02
    Negative Logits
     bedrock
    -0.67
     conceal
    -0.63
     discouraged
    -0.62
     cryptoc
    -0.62
     recl
    -0.62
     reinvest
    -0.60
     Hutchinson
    -0.60
     nonviolent
    -0.59
     renewables
    -0.59
     flip
    -0.58
    POSITIVE LOGITS
    âĢ
    1.38
    \.
    1.30
    »
    1.28
    ¨
    1.24
    *.
    1.22
    "(
    1.17
    _.
    1.16
    ãĢ
    1.16
    *,
    1.13
    ¯
    1.11
    Activations Density 0.209%

    No Known Activations