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. 10853
    Prev
    Next
    INDEX
    Explanations

    content related to updates, corrections, and metadata within articles

    oai_token-act-pair · gpt-3.5-turbo

    sequences of hyphens or dashes

    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
     Nasa
    -0.82
     scept
    -0.67
     Spartan
    -0.64
     migr
    -0.64
     organising
    -0.63
     SR
    -0.63
     Isis
    -0.63
     arte
    -0.62
     Nato
    -0.62
     Pillar
    -0.62
    POSITIVE LOGITS
    ---
    3.80
    ----
    2.89
    -----
    2.73
    ------
    2.59
    -------
    2.32
     ---
    2.31
    ---------
    2.18
    —-
    2.13
    --------
    2.04
    ----------
    1.97
    Activations Density 0.021%

    No Known Activations