Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 17384
    Prev
    Next
    INDEX
    Explanations

    words related to attributing or referencing content with a Creative Commons "BY" license

    oai_token-act-pair · gpt-3.5-turbo

    references to licensing or copyright attribution

    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
    val
    -0.63
     Gamb
    -0.62
     stabil
    -0.61
     oscill
    -0.61
    tracks
    -0.60
    est
    -0.60
    ths
    -0.60
    erson
    -0.58
    stall
    -0.57
     movements
    -0.57
    POSITIVE LOGITS
     BY
    3.64
    BY
    1.52
     WHEN
    1.06
     FROM
    1.05
     OF
    1.05
     WITH
    1.04
     AT
    1.04
     FOR
    1.00
     ON
    0.99
     AS
    0.98
    Activations Density 0.014%

    No Known Activations