© 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. 10-RES-JB
    11. 20216
    Prev
    Next
    INDEX
    Explanations

    locations and venues such as movie theaters and shopping malls

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks and their surrounding context, particularly periods and quotes

    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
     nodd
    -1.03
     quir
    -0.97
     caster
    -0.89
     tack
    -0.84
     metic
    -0.81
     exha
    -0.81
     favourite
    -0.79
     impro
    -0.79
     explan
    -0.77
     suspic
    -0.77
    POSITIVE LOGITS
     Together
    1.44
     Our
    1.37
     Unfortunately
    1.30
     Please
    1.30
    Together
    1.30
     Thank
    1.29
     Sadly
    1.27
     We
    1.26
    Tonight
    1.26
     Accordingly
    1.23
    Activations Density 0.486%

    No Known Activations