© 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. 9295
    Prev
    Next
    INDEX
    Explanations

    punctuation marks at the end of sentences

    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
    640
    +0.15
    0.7%
    373
    +0.12
    0.5%
    6
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.15
    0.08
    232
    +0.12
    0.06
    640
    +0.11
    0.06
    Negative Logits
    tesy
    -0.79
    itled
    -0.71
    itely
    -0.69
    itionally
    -0.69
    ciplinary
    -0.67
    ancies
    -0.67
    vice
    -0.66
    itional
    -0.65
    igent
    -0.65
     carbohyd
    -0.65
    POSITIVE LOGITS
     Anything
    1.25
     Whatever
    1.23
     Everywhere
    1.16
     And
    1.14
     Nope
    1.11
     Lots
    1.09
    etc
    1.08
     Literally
    1.07
     Repeat
    1.06
     Worse
    1.05
    Activations Density 0.256%

    No Known Activations