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. 2-RES-JB
    6. 24351
    Prev
    Next
    INDEX
    Explanations

    verbs associated with thinking or evaluating

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "consider" and related discussions about evaluating options or decisions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Jail
    -0.63
     Alley
    -0.62
     attendant
    -0.61
    Song
    -0.58
     afore
    -0.58
     injunction
    -0.57
    hell
    -0.57
     havoc
    -0.57
     pig
    -0.56
     Speedway
    -0.56
    POSITIVE LOGITS
    phas
    1.03
    ilitarian
    0.88
     consider
    0.86
    ibility
    0.85
    retion
    0.82
    aminer
    0.80
    prising
    0.78
    onyms
    0.76
    ate
    0.75
    cipled
    0.75
    Activations Density 0.017%

    No Known Activations