© 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. 24476
    Prev
    Next
    INDEX
    Explanations

    number values, such as dates, amounts, and quantities

    oai_token-act-pair · gpt-3.5-turbo

    key details related to events, such as dates and names associated with news stories

    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
     Whilst
    -1.12
     whilst
    -1.11
    avour
    -0.75
     Euros
    -0.74
     civilisation
    -0.72
     util
    -0.72
    shit
    -0.72
    ".[
    -0.71
     horrend
    -0.70
    !".
    -0.70
    POSITIVE LOGITS
     spokeswoman
    1.13
     statewide
    0.98
     spokesman
    0.97
     Legislature
    0.95
    CLOSE
    0.88
     superintendent
    0.85
     sheriff
    0.83
     aides
    0.83
     counselors
    0.82
     supervisors
    0.80
    Activations Density 1.463%

    No Known Activations