© 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 768
    9. 8-RES_FS768-JB
    11. 739
    Prev
    Next
    INDEX
    Explanations

    direct actions or impacts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    481
    +0.10
    0.5%
    515
    +0.09
    0.4%
    380
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.10
    0.20
    362
    +0.09
    0.15
    195
    +0.09
    0.06
    Negative Logits
    wright
    -0.88
    ingers
    -0.88
     Tycoon
    -0.86
     wisely
    -0.84
    girls
    -0.82
    mong
    -0.78
    horse
    -0.77
    girl
    -0.76
    imore
    -0.72
    rollers
    -0.71
    POSITIVE LOGITS
     amounts
    1.03
     exposure
    1.01
    ities
    0.99
     causal
    0.95
     consequences
    0.95
     ramifications
    0.93
     escalation
    0.93
     examination
    0.92
     consideration
    0.92
     repercussions
    0.92
    Activations Density 4.927%

    No Known Activations