INDEX

Explanations

words related to clothing or attire

oai_token-act-pair · gpt-3.5-turbo

references to clothing or attire

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

birth

-0.71

 Quincy

-0.69

 depress

-0.67

ollow

-0.65

elaide

-0.64

 gluc

-0.63

 Liqu

-0.62

ollo

-0.61

asper

-0.61

 Drama

-0.58

POSITIVE LOGITS

ters

0.98

 outfits

0.96

 outfit

0.95

 apparel

0.93

swick

0.91

puter

0.86

mentation

0.84

ments

0.84

glers

0.79

ema

0.79

Activations Density 0.012%

words related to clothing or attire

references to clothing or attire

No Comments

No Known Activations