© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1737
    Prev
    Next
    INDEX
    Explanations

    references to geoscience terms

    oai_token-act-pair · gpt-3.5-turbo

    terms related to geological science and geospatial concepts

    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
     pestic
    -0.73
     interim
    -0.72
     occas
    -0.71
     enactment
    -0.70
    ailability
    -0.69
     provisional
    -0.67
    ĪĴ
    -0.67
    EStreamFrame
    -0.64
     offence
    -0.63
    undai
    -0.62
    POSITIVE LOGITS
    cko
    0.75
    geist
    0.74
    Balt
    0.74
    odon
    0.73
    opter
    0.73
    acher
    0.71
    andise
    0.68
    ensen
    0.67
    achers
    0.67
    asma
    0.66
    Activations Density 0.071%

    No Known Activations