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. 22944
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences indicating an end or completion

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the word "over."

    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
    osity
    -0.75
     Forward
    -0.68
    resy
    -0.63
    ãĥı
    -0.63
     associates
    -0.62
     partName
    -0.61
    yssey
    -0.60
    Reference
    -0.59
     Galactic
    -0.58
     Readers
    -0.58
    POSITIVE LOGITS
    kill
    1.20
    blown
    1.16
    rated
    1.08
    priced
    1.06
    stated
    1.05
    reaching
    1.02
    whelming
    0.98
    valued
    0.97
    drive
    0.96
    loading
    0.94
    Activations Density 0.030%

    No Known Activations