Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 7-RES-JB
    6. 14937
    Prev
    Next
    INDEX
    Explanations

    references to historical figures or locations, specifically highlighting mentions of "Churchill."

    oai_token-act-pair · gpt-3.5-turbo

    references to historical figures, particularly Churchill and Gandhi

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    LOAD
    -0.93
     GOODMAN
    -0.85
    DEN
    -0.81
    ADRA
    -0.78
    ECTION
    -0.78
    lations
    -0.76
    lated
    -0.76
    lation
    -0.75
    parency
    -0.74
    tle
    -0.74
    POSITIVE LOGITS
     Jinn
    0.93
     Churchill
    0.91
    istically
    0.83
     Downs
    0.82
     Jr
    0.78
    reath
    0.76
    enstein
    0.75
     Chamberlain
    0.73
     Osw
    0.72
    weather
    0.71
    Activations Density 0.047%

    No Known Activations