© 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. 10-RES-JB
    11. 21348
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing a strong statement prefixed with "Never" or sentiments of caution

    oai_token-act-pair · gpt-3.5-turbo

    the repetition of the word "Never" and its variations in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    redients
    -0.71
    nikov
    -0.67
    iency
    -0.65
    ahime
    -0.64
    igans
    -0.63
    iant
    -0.61
    yk
    -0.61
    uliffe
    -0.61
    cil
    -0.60
    uers
    -0.60
    POSITIVE LOGITS
    theless
    1.75
    mind
    1.25
    winter
    1.11
     ceases
    1.03
     underestimate
    0.97
     mind
    0.95
     EVER
    0.84
    again
    0.84
     hesitate
    0.82
     ceased
    0.82
    Activations Density 0.035%

    No Known Activations