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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 8831
    Prev
    Next
    INDEX
    Explanations

    verbs relating to creation or achievement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    information related to sports statistics and player performance

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    heter
    -0.72
     Vest
    -0.66
     uterus
    -0.62
     dil
    -0.62
     womb
    -0.61
     herbs
    -0.61
     vest
    -0.60
    pac
    -0.60
    uld
    -0.58
     auditor
    -0.57
    POSITIVE LOGITS
    hift
    1.01
    ailable
    0.91
     sure
    0.82
     landfall
    0.80
    nces
    0.75
    ÄŁ
    0.74
    akable
    0.72
    asty
    0.71
    itures
    0.71
    ensions
    0.70
    Activations Density 0.070%

    No Known Activations