© 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. 6826
    Prev
    Next
    INDEX
    Explanations

    verbs associated with evaluating or judging a situation or outcome

    oai_token-act-pair · gpt-3.5-turbo

    repeated usage of the word "assess" in various contexts

    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
    stice
    -0.68
     Stories
    -0.68
     Wend
    -0.67
    ime
    -0.66
     happening
    -0.66
    stro
    -0.63
     disappearing
    -0.63
     thread
    -0.61
     dying
    -0.61
     Connection
    -0.61
    POSITIVE LOGITS
     assess
    3.24
     assessed
    2.28
     assessing
    2.14
     evaluate
    2.00
     evaluates
    1.64
     assessments
    1.62
     appra
    1.57
     analyse
    1.57
     assessment
    1.47
     quantify
    1.44
    Activations Density 0.019%

    No Known Activations