© 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. 0-RES-JB
    11. 16431
    Prev
    Next
    INDEX
    Explanations

    proper names, specifically related to the name "Frances"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rawler
    -0.67
     Nash
    -0.67
     bypass
    -0.65
     dunk
    -0.65
    awar
    -0.63
     Turing
    -0.63
    ript
    -0.62
     Bastard
    -0.61
    ,''
    -0.60
    Ĭ
    -0.60
    POSITIVE LOGITS
    Fr
    3.42
    Gi
    1.89
     Frances
    1.88
     Fr
    1.67
     Franc
    1.58
     Gi
    1.53
     Fra
    1.52
    Franc
    1.49
    Fran
    1.28
     Franks
    1.27
    Activations Density 0.020%

    No Known Activations