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. 10-RES-JB
    6. 21290
    Prev
    Next
    INDEX
    Explanations

    instances of the word "roll" followed by numbers, connoting a variety of contexts such as physical actions or procedural steps

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "roll" and its variations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ILE
    -0.76
    acters
    -0.69
    xual
    -0.68
    orial
    -0.63
    bub
    -0.63
    Hunt
    -0.63
    selage
    -0.62
    raints
    -0.62
    rent
    -0.62
    ulia
    -0.61
    POSITIVE LOGITS
    icking
    0.92
     boulder
    0.86
    back
    0.86
     dice
    0.84
    outs
    0.77
    anut
    0.77
     hills
    0.76
     downhill
    0.76
    ahon
    0.76
     oats
    0.75
    Activations Density 0.039%

    No Known Activations