© 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. 10-RES-JB
    11. 16286
    Prev
    Next
    INDEX
    Explanations

    statements related to company announcements

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of specific symbols or punctuation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     guy
    -0.88
     asshole
    -0.83
     mathemat
    -0.83
     reflex
    -0.82
     favour
    -0.81
     overth
    -0.81
     phenomen
    -0.80
     dude
    -0.79
     subconscious
    -0.79
     manif
    -0.78
    POSITIVE LOGITS
    Currently
    1.29
    Additionally
    1.28
    Previously
    1.27
    ccording
    1.27
    Explore
    1.24
    Based
    1.24
    Additional
    1.21
    Experts
    1.21
    Due
    1.19
    Reporting
    1.18
    Activations Density 0.160%

    No Known Activations