© 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. 18796
    Prev
    Next
    INDEX
    Explanations

    sentences containing the pronoun "it" followed by the verb "be" or related verbs like "seem" or "appear", possibly describing evaluations or observations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "It" often indicating statements and opinions

    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
     Peninsula
    -0.63
     Eighth
    -0.61
    Had
    -0.61
    ometown
    -0.58
     Meadow
    -0.57
    orthern
    -0.56
    IDE
    -0.54
     Entered
    -0.54
     Bunker
    -0.53
    IDA
    -0.53
    POSITIVE LOGITS
     varies
    1.26
     depends
    1.22
     helps
    1.20
     takes
    1.16
     happens
    1.11
     boils
    1.11
     tends
    1.11
     doesnt
    1.11
    unes
    1.10
     becomes
    1.09
    Activations Density 0.181%

    No Known Activations