© 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. 11670
    Prev
    Next
    INDEX
    Explanations

    references to surveys conducted by geological organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to surveys and survey-related data

    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
    redo
    -0.81
     innocence
    -0.70
     oper
    -0.67
     nood
    -0.65
     bud
    -0.65
    abl
    -0.63
     endings
    -0.63
     amput
    -0.63
     incarn
    -0.63
     Romero
    -0.62
    POSITIVE LOGITS
     Survey
    3.99
     Surve
    2.85
     survey
    1.94
     surveys
    1.82
     Census
    1.45
     surveyed
    1.44
     Study
    1.37
     surve
    1.23
     Surveillance
    1.12
     Examination
    1.10
    Activations Density 0.009%

    No Known Activations