© 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. 10-RES-JB
    11. 14045
    Prev
    Next
    INDEX
    Explanations

    references to specific entities or objects like "Baby Pokémon," "partner," "opponent," "dog," "device," "aircraft," "computer," and "patient."

    oai_token-act-pair · gpt-3.5-turbo

    references to characters, animals, and entities in specific contexts or scenarios

    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
    DragonMagazine
    -1.06
    ntil
    -0.84
    TPPStreamerBot
    -0.70
    ranging
    -0.69
    soDeliveryDate
    -0.69
    udeb
    -0.68
    ©¶æ¥µ
    -0.67
    Dub
    -0.65
    anmar
    -0.64
    externalActionCode
    -0.63
    POSITIVE LOGITS
     exceeds
    1.57
     fails
    1.48
     reaches
    1.44
     enters
    1.40
     succeeds
    1.39
     disagrees
    1.34
     detects
    1.33
     receives
    1.33
     engages
    1.32
     gets
    1.29
    Activations Density 0.293%

    No Known Activations