© 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 1536
    9. 8-RES_FS1536-JB
    11. 54
    Prev
    Next
    INDEX
    Explanations

    personal experiences or descriptions in a narrative format

    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
    446
    +0.11
    0.5%
    489
    +0.08
    0.4%
    630
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.11
    0.12
    477
    +0.08
    0.07
    139
    +0.08
    0.09
    Negative Logits
    WER
    -0.72
    <?
    -0.67
    never
    -0.66
    oner
    -0.64
     Constantin
    -0.64
     Engineers
    -0.62
     Cosponsors
    -0.60
     aliases
    -0.60
    bold
    -0.60
    ¶ħ
    -0.59
    POSITIVE LOGITS
     arrives
    0.95
     arrive
    0.92
     expires
    0.90
     arrived
    0.86
     enters
    0.83
     reaches
    0.79
     finished
    0.77
     completes
    0.76
     toured
    0.74
     transitioned
    0.74
    Activations Density 3.015%

    No Known Activations