© 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. 0-RES-JB
    11. 11545
    Prev
    Next
    INDEX
    Explanations

    weights in pounds

    oai_token-act-pair · gpt-3.5-turbo

    units of measurement related to weight

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ways
    -0.73
     patches
    -0.72
     graffiti
    -0.70
     mosaic
    -0.66
     Nost
    -0.65
     Dominion
    -0.65
     conserv
    -0.62
     tragedies
    -0.61
     patch
    -0.61
     reinvest
    -0.61
    POSITIVE LOGITS
    lb
    3.26
    lbs
    2.98
     lb
    2.97
     lbs
    2.80
    kg
    1.37
    LB
    1.29
     kg
    1.14
    lua
    1.08
    pound
    1.05
     pounds
    1.04
    Activations Density 0.017%

    No Known Activations