© 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. 886
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions involving multiple people working together

    oai_token-act-pair · gpt-3.5-turbo

    verbs associated with persuasion and interactions within groups

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ukong
    -0.60
    ãĤ¼
    -0.57
    isin
    -0.55
    ftime
    -0.53
    ocent
    -0.52
    ¯
    -0.49
    mosp
    -0.49
    >>\
    -0.49
    inventoryQuantity
    -0.49
     Earthquake
    -0.49
    POSITIVE LOGITS
     them
    1.32
     ones
    1.24
     theirs
    1.17
     THEM
    1.14
    them
    1.08
     apiece
    1.08
     Them
    1.07
    They
    0.97
    they
    0.97
     they
    0.89
    Activations Density 1.849%

    No Known Activations