© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 15353
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or events involving breaking up something

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to breaking or separating

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Hunts
    -0.63
    ait
    -0.62
    WHERE
    -0.62
     projecting
    -0.61
     Boxing
    -0.61
    gee
    -0.60
    tto
    -0.59
    velop
    -0.59
    READ
    -0.59
    anwhile
    -0.58
    POSITIVE LOGITS
    river
    0.95
    arity
    0.81
    rates
    0.78
    dates
    0.75
    glomer
    0.74
    stocks
    0.73
    rate
    0.73
    scenes
    0.72
    stock
    0.70
    rations
    0.68
    Activations Density 0.028%

    No Known Activations