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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 8-RES-JB
    6. 5871
    Prev
    Next
    INDEX
    Explanations

    phrases indicating individual items or entities within a broader context

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Each"

    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
    abs
    -0.79
    tops
    -0.74
    rovers
    -0.73
    haven
    -0.72
    dn
    -0.69
    duc
    -0.68
    Chain
    -0.66
    children
    -0.66
    ables
    -0.65
    stones
    -0.65
    POSITIVE LOGITS
     successive
    1.19
     iteration
    0.98
     individual
    0.88
     participant
    0.87
     person
    0.84
    where
    0.83
     month
    0.83
     piece
    0.81
     member
    0.80
     year
    0.78
    Activations Density 0.038%

    No Known Activations