Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 951
    Prev
    Next
    INDEX
    Explanations

    time-related information such as dates and times

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    373
    +0.17
    0.8%
    497
    +0.09
    0.4%
    18
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.17
    -0.02
    403
    +0.09
    0.06
    176
    +0.09
    0.04
    Negative Logits
     arms
    -0.60
     expansion
    -0.58
     "{
    -0.57
     Abrams
    -0.57
     Viz
    -0.56
     pudding
    -0.55
    ospace
    -0.55
     landsl
    -0.55
     pregnancy
    -0.53
     abuser
    -0.53
    POSITIVE LOGITS
    00
    1.44
    30
    1.39
    05
    1.38
    09
    1.38
    08
    1.37
    06
    1.37
    04
    1.36
    07
    1.35
    59
    1.35
    02
    1.31
    Activations Density 0.780%

    No Known Activations