© 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. 23564
    Prev
    Next
    INDEX
    Explanations

    terms related to technology companies and products

    oai_token-act-pair · gpt-3.5-turbo

    chemical compounds and related terms

    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
     lightly
    -0.58
     sufficiently
    -0.53
     Patreon
    -0.53
     Dangerous
    -0.53
     Democr
    -0.52
     convenient
    -0.51
     abruptly
    -0.51
    pects
    -0.50
     harshly
    -0.49
    .--
    -0.48
    POSITIVE LOGITS
     etc
    1.10
    etc
    0.92
    âĢ¦)
    0.73
    whatever
    0.72
    rote
    0.65
    cussion
    0.62
    ...)
    0.55
    ife
    0.55
    ItemThumbnailImage
    0.55
     TBA
    0.54
    Activations Density 0.607%

    No Known Activations