© 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. 10231
    Prev
    Next
    INDEX
    Explanations

    titles of books

    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
    557
    +0.11
    0.5%
    329
    +0.10
    0.4%
    176
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.11
    0.03
    308
    +0.10
    0.03
    744
    +0.09
    0.03
    Negative Logits
    esters
    -0.68
     base
    -0.67
     bulb
    -0.64
     fanbase
    -0.64
     joke
    -0.64
     microphone
    -0.64
     performer
    -0.63
     wink
    -0.63
     teammate
    -0.63
     assist
    -0.63
    POSITIVE LOGITS
     Making
    1.11
     Lessons
    1.08
     Principles
    1.07
     Handbook
    1.06
     Understanding
    1.06
     Reasons
    1.02
     Respons
    0.99
     Changes
    0.98
     Towards
    0.97
     Bringing
    0.95
    Activations Density 0.303%

    No Known Activations