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. 10-RES-JB
    6. 11842
    Prev
    Next
    INDEX
    Explanations

    superlatives or intense descriptions regarding qualities or actions

    oai_token-act-pair · gpt-3.5-turbo

    superlative claims or descriptors

    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
    osponsors
    -0.91
    ĸļ士
    -0.81
    rers
    -0.80
    azard
    -0.79
    ordes
    -0.76
    legates
    -0.75
    renheit
    -0.74
    folios
    -0.74
    ptives
    -0.74
    ilings
    -0.73
    POSITIVE LOGITS
     thing
    1.86
     aspect
    1.35
     piece
    1.26
     moment
    1.26
    THING
    1.25
    thing
    1.24
     reason
    1.23
     topic
    1.19
     item
    1.19
     part
    1.11
    Activations Density 0.327%

    No Known Activations