© 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. 3-RES-JB
    11. 8343
    Prev
    Next
    INDEX
    Explanations

    phrases related to revealing or exposing

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate exposure or revelation of underlying issues or truths

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Pic
    -0.60
    Lower
    -0.59
    TX
    -0.58
     pleading
    -0.57
    ford
    -0.57
     grade
    -0.56
    chan
    -0.55
    ãĥ¬
    -0.55
     propensity
    -0.55
    Chart
    -0.54
    POSITIVE LOGITS
    */(
    0.73
    umbn
    0.69
    utters
    0.67
    roads
    0.66
    apeake
    0.66
    clusions
    0.65
     runway
    0.64
    ordinate
    0.62
     ashore
    0.60
    htaking
    0.60
    Activations Density 0.561%

    No Known Activations