INDEX

Explanations

implicit or indirect statements or suggestions

oai_token-act-pair · gpt-3.5-turbo

phrases related to implications or suggestions

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

 Reds

-0.78

mir

-0.74

thumbnails

-0.74

Ern

-0.67

anke

-0.65

gard

-0.64

unker

-0.64

dan

-0.62

home

-0.61

HCR

-0.61

POSITIVE LOGITS

 imply

0.91

 implied

0.85

 infer

0.77

 WARRANT

0.76

antle

0.73

 guiActiveUn

0.73

 endorsement

0.73

icit

0.71

 coupling

0.71

heny

0.71

Activations Density 0.029%

implicit or indirect statements or suggestions

phrases related to implications or suggestions

No Comments

No Known Activations