© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 5680
    Prev
    Next
    INDEX
    Explanations

    phrases indicating interpretation or perception of events or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating perception or interpretation of events

    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
    enne
    -0.64
    luaj
    -0.63
    hung
    -0.62
    +,
    -0.60
    ichen
    -0.60
     Doesn
    -0.59
     prefers
    -0.58
     varies
    -0.58
    Initialized
    -0.57
    inately
    -0.57
    POSITIVE LOGITS
     testament
    1.14
     confirmation
    1.12
     proof
    1.10
     retaliation
    1.08
     sympt
    1.07
     validation
    1.04
     indicative
    0.99
     vind
    0.99
    evidence
    0.98
     paving
    0.95
    Activations Density 0.253%

    No Known Activations