© 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. 14036
    Prev
    Next
    INDEX
    Explanations

    quotes or phrases in a question-response format

    oai_token-act-pair · gpt-3.5-turbo

    question marks and conversational cues indicating uncertainty or requests for confirmation

    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
     multiplied
    -0.78
     fanc
    -0.77
     fleeing
    -0.77
     trave
    -0.75
     chants
    -0.75
     harassing
    -0.75
     pict
    -0.74
     stray
    -0.74
     migr
    -0.74
     forgotten
    -0.73
    POSITIVE LOGITS
    JM
    1.46
    JB
    1.41
    Answer
    1.37
    JV
    1.36
    RH
    1.32
    EH
    1.32
    MH
    1.29
    JP
    1.26
    DW
    1.26
    JS
    1.25
    Activations Density 0.069%

    No Known Activations