© 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. 6189
    Prev
    Next
    INDEX
    Explanations

    words related to correctness, agreement, or justification in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    adverbs expressing correctness or justification

    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
    contact
    -0.75
     Alternate
    -0.63
     supplemental
    -0.58
     alternate
    -0.56
    hide
    -0.56
    olution
    -0.56
     plan
    -0.56
    azines
    -0.56
    rawdownloadcloneembedreportprint
    -0.56
     contact
    -0.55
    POSITIVE LOGITS
     rightly
    3.60
     rightfully
    3.11
     understandably
    2.31
     wrongly
    1.72
     wisely
    1.49
     unsurprisingly
    1.44
     doubtless
    1.36
     errone
    1.34
     legitimately
    1.32
     mistakenly
    1.26
    Activations Density 0.025%

    No Known Activations