© 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 12288
    9. 8-RES_FS12288-JB
    11. 1450
    Prev
    Next
    INDEX
    Explanations

    phrases related to predictions or estimates of future trends or values

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    105
    +0.12
    0.6%
    139
    +0.12
    0.5%
    665
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.12
    0.05
    665
    +0.12
    0.05
    551
    +0.11
    0.05
    Negative Logits
     Sure
    -0.71
     imagining
    -0.66
    checking
    -0.65
    amps
    -0.65
    76561
    -0.64
     Checking
    -0.64
     Connection
    -0.62
     Thinking
    -0.61
     Promotion
    -0.60
     Explan
    -0.60
    POSITIVE LOGITS
     be
    1.15
     arrive
    1.12
     continue
    1.09
     intensify
    1.00
     settle
    1.00
     become
    0.99
     prevail
    0.98
     remain
    0.97
     reap
    0.97
     yield
    0.95
    Activations Density 0.185%

    No Known Activations