© 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. 14136
    Prev
    Next
    INDEX
    Explanations

    sentences containing the word "a" followed by an adjective

    oai_token-act-pair · gpt-3.5-turbo

    phrases containing the article "a" followed by descriptors or nouns, suggesting evaluation or categorization

    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
    ographs
    -0.79
    units
    -0.76
    ansson
    -0.75
    effects
    -0.75
    aday
    -0.72
    itely
    -0.70
    Edit
    -0.70
    Element
    -0.69
     arcs
    -0.67
    iffs
    -0.67
    POSITIVE LOGITS
     mistake
    1.31
     violation
    1.19
     hoax
    1.14
     ploy
    1.11
     nuisance
    1.10
     distraction
    1.09
     joke
    1.09
     betrayal
    1.08
     waste
    1.06
     failure
    1.06
    Activations Density 0.219%

    No Known Activations