© 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. 10115
    Prev
    Next
    INDEX
    Explanations

    prepositions followed by action verbs

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "it's hard to."

    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
    ELD
    -0.62
     courtesy
    -0.61
     Reborn
    -0.61
     reperto
    -0.60
     Era
    -0.59
    loads
    -0.58
    furt
    -0.58
     Glob
    -0.58
     Creat
    -0.57
    borough
    -0.57
    POSITIVE LOGITS
     imagine
    1.32
     reconcile
    1.21
     conceive
    1.20
     quantify
    1.19
     argue
    1.12
     envision
    1.11
     overest
    1.10
     believe
    1.05
     deny
    1.03
     comprehend
    1.03
    Activations Density 0.046%

    No Known Activations