© 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. 18898
    Prev
    Next
    INDEX
    Explanations

    terms related to technology and data storage methods

    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
    698
    +0.11
    0.5%
    506
    +0.11
    0.5%
    546
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    546
    +0.11
    0.06
    43
    +0.11
    0.04
    73
    +0.10
    0.04
    Negative Logits
    escription
    -0.76
    lift
    -0.64
     behalf
    -0.63
     grievances
    -0.61
    ore
    -0.60
     Compliance
    -0.58
    atives
    -0.57
    vice
    -0.56
    SHIP
    -0.56
    isure
    -0.56
    POSITIVE LOGITS
     except
    0.88
    depending
    0.84
     although
    0.80
    although
    0.76
     albeit
    0.72
     nor
    0.72
     depending
    0.71
    TPPStreamerBot
    0.69
     insofar
    0.69
     Flavoring
    0.68
    Activations Density 0.292%

    No Known Activations