© 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. 1-RES-JB
    11. 8573
    Prev
    Next
    INDEX
    Explanations

    past tense verbs

    oai_token-act-pair · gpt-3.5-turbo

    the verb "was" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Layer 1: Is/Was tokens
    was (but much more strongly if was appears previously in the text)

    No Comments

    Negative Logits
    inav
    -0.75
    eers
    -0.62
     Extend
    -0.60
     Which
    -0.58
     Suit
    -0.58
     Moines
    -0.57
    arming
    -0.56
     Individual
    -0.55
     Chel
    -0.54
     Highlands
    -0.54
    POSITIVE LOGITS
    nt
    0.96
    hes
    0.87
     definitely
    0.85
    hers
    0.83
    wolf
    0.78
    wolves
    0.76
     supposed
    0.76
     certainly
    0.74
    rael
    0.72
     also
    0.71
    Activations Density 0.363%

    No Known Activations