© 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. 0-RES-JB
    11. 4758
    Prev
    Next
    INDEX
    Explanations

    mentions of a specific programming language

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the programming language Ruby and its related frameworks

    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
     monop
    -0.84
    EH
    -0.73
    minist
    -0.68
    ���
    -0.67
     fined
    -0.65
    Pac
    -0.63
    ��
    -0.63
     plur
    -0.63
     prob
    -0.62
     dispens
    -0.61
    POSITIVE LOGITS
     Ruby
    3.73
    Ruby
    3.33
     ruby
    2.38
    ruby
    2.16
     Rails
    1.79
     Pyrrha
    1.75
     Jaune
    1.69
     Sapphire
    1.69
     Weiss
    1.67
    Weiss
    1.59
    Activations Density 0.019%

    No Known Activations