© 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. 1544
    Prev
    Next
    INDEX
    Explanations

    instructions or technical details related to file browsers, file transfer modes, and message processing systems

    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
    447
    +0.13
    0.6%
    231
    +0.12
    0.5%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.07
    225
    +0.12
    0.05
    519
    +0.10
    0.04
    Negative Logits
    heit
    -0.72
    abama
    -0.68
    Rail
    -0.68
    DEF
    -0.68
    ahime
    -0.68
    bush
    -0.64
    DAY
    -0.64
    BUR
    -0.63
    fml
    -0.63
    Benz
    -0.62
    POSITIVE LOGITS
     extraord
    0.84
     puzz
    0.80
    itiz
    0.71
    etting
    0.70
     format
    0.69
    ror
    0.68
    orer
    0.67
    ophon
    0.66
    etter
    0.66
    rors
    0.66
    Activations Density 0.338%

    No Known Activations