INDEX

Explanations

phrases indicating difficulty or challenge

oai_token-act-pair · gpt-3.5-turbo

phrases indicating difficulty or ease in performing tasks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

afety

-0.81

erity

-0.75

DragonMagazine

-0.63

 Anthem

-0.62

 Nationwide

-0.61

 Past

-0.60

Bet

-0.60

forth

-0.60

ERG

-0.59

 Stories

-0.59

POSITIVE LOGITS

 obtain

1.07

 attain

1.03

 emulate

1.02

 achieve

1.00

 navigate

0.98

 quantify

0.97

 acquire

0.96

 imitate

0.95

 comprehend

0.94

 maintain

0.94

Activations Density 0.083%

phrases indicating difficulty or challenge

phrases indicating difficulty or ease in performing tasks

No Comments

No Known Activations