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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 0-RES-JB
    6. 8888
    Prev
    Next
    INDEX
    Explanations

    the word "ref" or its variations

    oai_token-act-pair · gpt-3.5-turbo

    references to regulations or legal frameworks related to the refining industry

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hare
    -0.89
     Sora
    -0.76
    conom
    -0.71
     Charm
    -0.70
    arious
    -0.67
    chan
    -0.67
    hent
    -0.67
    geist
    -0.66
    gamer
    -0.66
    achine
    -0.65
    POSITIVE LOGITS
     ref
    3.88
     Ref
    2.18
    ref
    1.84
    Ref
    1.83
     refin
    1.42
    REF
    1.34
     refill
    1.33
     refining
    1.32
     refuel
    1.17
     cond
    1.15
    Activations Density 0.004%

    No Known Activations