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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 8-RES-JB
    6. 18863
    Prev
    Next
    INDEX
    Explanations

    details and descriptions related to crime and investigations

    oai_token-act-pair · gpt-3.5-turbo

    context related to societal issues and community challenges

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     RNC
    -0.72
     neocons
    -0.72
     Comey
    -0.69
     rhetorical
    -0.67
     Kislyak
    -0.67
     CBO
    -0.67
     Tillerson
    -0.65
     Mattis
    -0.65
     Politico
    -0.64
    Trump
    -0.63
    POSITIVE LOGITS
     Guinness
    0.89
     civilisation
    0.87
     organisers
    0.78
     Forestry
    0.74
     forestry
    0.73
     villagers
    0.71
     licences
    0.70
    mercial
    0.69
     UNESCO
    0.68
     practise
    0.67
    Activations Density 1.853%

    No Known Activations