© 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. 8-RES-JB
    11. 22643
    Prev
    Next
    INDEX
    Explanations

    mentions of the chemical element iodine

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "io"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rier
    -0.75
    ancies
    -0.67
    holder
    -0.67
    ishing
    -0.66
     complimentary
    -0.64
    ished
    -0.62
    uates
    -0.61
    rap
    -0.61
     Compass
    -0.60
    ivities
    -0.60
    POSITIVE LOGITS
    ppo
    1.18
    hazard
    1.00
    ctl
    0.97
    ffic
    0.96
    active
    0.96
    cean
    0.96
    ÄŁ
    0.90
    pec
    0.87
    lex
    0.87
    xit
    0.85
    Activations Density 0.033%

    No Known Activations