Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 21002
    Prev
    Next
    INDEX
    Explanations

    contractions with the word 'is' and 'has'

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the contraction "it's."

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

    No Comments

    Negative Logits
    scope
    -0.61
    velop
    -0.61
     este
    -0.60
     Deaths
    -0.59
    rence
    -0.58
    umbnail
    -0.58
    911
    -0.58
    rating
    -0.58
    ESE
    -0.58
    ={
    -0.57
    POSITIVE LOGITS
     gonna
    1.11
     gotta
    1.07
     been
    0.89
     unclear
    0.84
     worth
    0.84
     got
    0.82
     impossible
    0.79
    Ĥİ
    0.79
     not
    0.77
    etsk
    0.77
    Activations Density 0.089%

    No Known Activations