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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 6194
    Prev
    Next
    INDEX
    Explanations

    years and dates, particularly with historical or notable events

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "and" indicating connections or continuations in statements

    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
    nikov
    -0.70
    BSD
    -0.64
     Generic
    -0.63
    ellectual
    -0.63
    Development
    -0.63
    onica
    -0.60
    natureconservancy
    -0.60
     Mechdragon
    -0.59
    unts
    -0.58
    icated
    -0.58
    POSITIVE LOGITS
     thereafter
    1.13
     onwards
    1.02
     proceeded
    0.97
     lasted
    0.96
     subsequently
    0.96
     then
    0.95
     thus
    0.92
     onward
    0.92
     culminated
    0.91
     consequently
    0.90
    Activations Density 0.164%

    No Known Activations