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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24221
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to forest locations

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Forest" and related terms

    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
    multipl
    -0.74
    ollah
    -0.74
     recip
    -0.71
    ÏĢ
    -0.70
     á
    -0.69
     transmit
    -0.68
    iframe
    -0.67
     iP
    -0.67
     pc
    -0.66
    á
    -0.66
    POSITIVE LOGITS
     Forest
    3.93
    Forest
    2.63
     forest
    2.11
    forest
    2.10
     forests
    1.85
     Forestry
    1.78
     forestry
    1.55
     Forrest
    1.47
     Timber
    1.39
     Tree
    1.31
    Activations Density 0.010%

    No Known Activations