© 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. 11-RES-JB
    11. 7066
    Prev
    Next
    INDEX
    Explanations

    adverbs or adjectives related to forceful or coercive actions

    oai_token-act-pair · gpt-3.5-turbo

    terms related to coercive actions and legality issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ERAL
    -0.73
     naturally
    -0.71
     inevitably
    -0.66
    formance
    -0.66
    Liter
    -0.64
     Sight
    -0.64
     insistence
    -0.63
     willingness
    -0.63
    erer
    -0.62
     enthus
    -0.62
    POSITIVE LOGITS
     removed
    1.06
     transported
    1.05
     evacuated
    1.04
     transitioned
    1.03
     prevented
    1.02
     executed
    1.02
     converted
    1.02
     extracted
    1.00
     detained
    1.00
     dismantled
    1.00
    Activations Density 0.134%

    No Known Activations