© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 19964
    Prev
    Next
    INDEX
    Explanations

    references to pizza

    oai_token-act-pair · gpt-3.5-turbo

    references to pizza and related food items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Scorpion
    -0.78
     Battery
    -0.73
    ume
    -0.71
     Spot
    -0.71
     bracket
    -0.69
    bolt
    -0.68
    icone
    -0.67
     Bolt
    -0.67
    thia
    -0.67
    Shot
    -0.66
    POSITIVE LOGITS
     pizz
    2.36
     paed
    1.49
     pedoph
    1.39
     coff
    1.33
     Gork
    1.19
     clos
    1.17
     iceberg
    1.14
    ér
    1.14
     ambul
    1.12
     pies
    1.11
    Activations Density 0.069%

    No Known Activations