Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 9776
    Prev
    Next
    INDEX
    Explanations

    words indicating the need to perform or fulfill a certain action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "had to" in various contexts

    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
     Nights
    -0.64
     Ideal
    -0.64
     Oval
    -0.64
     Emerging
    -0.63
    AAA
    -0.61
     Lif
    -0.60
     Principles
    -0.60
     Dominion
    -0.59
    Community
    -0.59
    DEM
    -0.58
    POSITIVE LOGITS
     endure
    1.28
     contend
    1.23
     rely
    1.19
     fend
    1.19
     resort
    1.13
     wait
    1.12
     cope
    1.10
     admit
    1.10
     scramble
    1.10
     deal
    1.07
    Activations Density 0.102%

    No Known Activations