INDEX

Explanations

references or citations in text

oai_token-act-pair · gpt-3.5-turbo

references to the act of citing sources or evidence

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orld

-0.81

ipeg

-0.74

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.74

xy

-0.71

hattan

-0.69

ibaba

-0.68

visors

-0.67

ascript

-0.67

amination

-0.66

ixtape

-0.66

POSITIVE LOGITS

 citations

0.95

 citation

0.87

enza

0.81

 cite

0.78

 cited

0.76

 citing

0.75

âĨĳ

0.70

 warnings

0.70

Cit

0.70

 Forbidden

0.70

Activations Density 0.019%

references or citations in text

references to the act of citing sources or evidence

No Comments

No Known Activations