© 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. 1-RES-JB
    11. 20881
    Prev
    Next
    INDEX
    Explanations

    terms related to mechanical operations or tools, specifically focusing on parts of a machine

    oai_token-act-pair · gpt-3.5-turbo

    terms related to cycling and mechanisms of motion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aldo
    -1.00
    uality
    -0.90
    uary
    -0.85
    omics
    -0.85
    ework
    -0.85
    icle
    -0.83
    icles
    -0.81
    obyl
    -0.78
    ional
    -0.77
    yk
    -0.76
    POSITIVE LOGITS
     pedal
    0.84
    NK
    0.83
     pedals
    0.82
    ptin
    0.76
    Charl
    0.73
     Harley
    0.70
    WAYS
    0.68
    weed
    0.68
    pal
    0.67
     Pwr
    0.67
    Activations Density 0.029%

    No Known Activations