© 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. 1690
    Prev
    Next
    INDEX
    Explanations

    words related to appreciation and feedback

    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
    186
    +0.10
    0.5%
    276
    +0.10
    0.5%
    168
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    186
    +0.10
    0.04
    168
    +0.10
    0.04
    276
    +0.10
    0.04
    Negative Logits
     Strait
    -0.68
     Archdemon
    -0.65
    arij
    -0.64
    ONSORED
    -0.61
    natureconservancy
    -0.57
     sshd
    -0.57
    osc
    -0.57
    ãĤ£
    -0.57
     digestion
    -0.57
    Edge
    -0.57
    POSITIVE LOGITS
    abouts
    1.18
    tical
    1.15
    tics
    1.14
    tic
    0.92
     Comes
    0.89
    after
    0.82
     goes
    0.80
    with
    0.78
    upon
    0.76
    ford
    0.73
    Activations Density 0.046%

    No Known Activations