© 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. 9865
    Prev
    Next
    INDEX
    Explanations

    statistics or numerical patterns, particularly with a focus on the frequency of occurrences

    oai_token-act-pair · gpt-3.5-turbo

    numbers and statistical references, particularly those phrased as "one in X."

    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
    photos
    -0.69
    ilts
    -0.61
    steps
    -0.60
    thouse
    -0.57
     Countdown
    -0.56
     Dynam
    -0.55
     Original
    -0.55
    endars
    -0.55
    events
    -0.54
     Bounce
    -0.54
    POSITIVE LOGITS
     thirds
    1.40
    thirds
    1.15
     tenth
    1.06
     percent
    1.05
     quarter
    1.02
     hundred
    1.01
     million
    1.00
     trillion
    0.99
     billion
    0.98
     fifth
    0.97
    Activations Density 0.131%

    No Known Activations