Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 5850
    Prev
    Next
    INDEX
    Explanations

    positive and negative adjectives

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that denote positive or negative evaluations

    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
    steps
    -0.89
    ques
    -0.83
    busters
    -0.77
     Measures
    -0.76
    hops
    -0.76
    iddles
    -0.75
    ateurs
    -0.75
     breaches
    -0.74
    changes
    -0.74
    Nazis
    -0.74
    POSITIVE LOGITS
     knack
    1.30
     tendency
    1.29
     penchant
    1.28
     reputation
    1.23
     relationship
    1.10
     grasp
    1.09
     inclination
    1.03
     foothold
    1.03
     outlook
    1.03
     propensity
    1.03
    Activations Density 0.272%

    No Known Activations