© 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. 11-RES-JB
    11. 21229
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial or economic situations

    oai_token-act-pair · gpt-3.5-turbo

    end punctuation marks in a variety of contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     dimensional
    -0.91
     deity
    -0.90
     gobl
    -0.90
    ŃĶ
    -0.87
     tremend
    -0.84
     thous
    -0.84
     defe
    -0.83
     tradem
    -0.83
     uncond
    -0.82
     mate
    -0.81
    POSITIVE LOGITS
     Experts
    1.44
     Many
    1.36
     Estimates
    1.35
     Increasing
    1.35
     Across
    1.33
     Millions
    1.33
     Researchers
    1.32
     Hundreds
    1.31
     Nearly
    1.30
     Some
    1.29
    Activations Density 0.406%

    No Known Activations