© 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. 16680
    Prev
    Next
    INDEX
    Explanations

    verbs involving action or behavior

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the word "do" across 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
     Methods
    -0.65
     simultane
    -0.64
     Prosecutor
    -0.59
     Hurricanes
    -0.58
     Citation
    -0.57
     Commissioner
    -0.55
     champagne
    -0.54
    rification
    -0.53
     Legions
    -0.53
     WATCHED
    -0.53
    POSITIVE LOGITS
    omsday
    1.38
    ppel
    1.37
    ozy
    1.25
    gging
    1.25
    herty
    1.23
    ggie
    1.18
    ggy
    1.16
    berman
    1.15
    ctors
    1.13
    orman
    1.13
    Activations Density 0.048%

    No Known Activations