© 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. 16468
    Prev
    Next
    INDEX
    Explanations

    references to a product or brand named "Legacy"

    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
    iour
    -0.76
     Beaver
    -0.74
     ther
    -0.73
     ¶
    -0.72
     Gad
    -0.70
     gu
    -0.69
     mete
    -0.67
     Fernand
    -0.67
     sack
    -0.66
     Moz
    -0.66
    POSITIVE LOGITS
     Legacy
    3.21
     Vintage
    1.33
     Eternal
    1.26
     Mono
    1.11
     Origins
    1.09
     Decay
    1.03
    span
    0.93
     Esper
    0.93
     Nicol
    0.92
    deck
    0.92
    Activations Density 0.040%

    No Known Activations