© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 40435
    Prev
    Next
    INDEX
    Explanations

    concepts or ideas that have been thought of or created

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    612
    +0.09
    0.5%
    339
    +0.09
    0.5%
    630
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.09
    0.08
    195
    +0.09
    -0.02
    541
    +0.08
    0.09
    Negative Logits
    ammy
    -0.56
    anca
    -0.53
    anye
    -0.51
    anwhile
    -0.51
     reinforcement
    -0.50
     ubiqu
    -0.49
    upon
    -0.49
     amy
    -0.49
    acca
    -0.49
     intimidation
    -0.49
    POSITIVE LOGITS
    smanship
    0.71
    REAM
    0.71
    JECT
    0.65
     Bout
    0.64
    ãĥĩãĤ£
    0.63
    eers
    0.63
     notions
    0.62
    iour
    0.61
     posit
    0.58
    agram
    0.58
    Activations Density 11.019%

    No Known Activations