© 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. 8-RES-JB
    11. 4945
    Prev
    Next
    INDEX
    Explanations

    specific terms related to publications or literature such as book titles, authors, and references

    oai_token-act-pair · gpt-3.5-turbo

    reference to changes or significant events related to time and context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •contextual features
    fires strongly on common phrases like "that it", "that the" "they do". Fires individually on " they", " one", " their"

    No Comments

    Negative Logits
     Seym
    -0.76
    DAQ
    -0.67
    Gall
    -0.65
     Moroc
    -0.65
     Berg
    -0.64
     Niet
    -0.63
     Fern
    -0.63
    Finish
    -0.62
    ãĥ¼ãĥ³
    -0.62
     Jagu
    -0.61
    POSITIVE LOGITS
     retarded
    0.71
     unders
    0.71
    ';
    0.65
    nces
    0.62
     plurality
    0.60
     ye
    0.60
     discriminate
    0.60
     persons
    0.60
    ignty
    0.58
     discrim
    0.57
    Activations Density 0.461%

    No Known Activations