INDEX

Explanations

specific mentions of the word "set."

oai_token-act-pair · gpt-3.5-turbo

references to a collection or group of items

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

IBLE

-0.80

ãĤĭ

-0.72

ibly

-0.71

issance

-0.70

antly

-0.66

ï¸ı

-0.65

 Taste

-0.61

anced

-0.61

perse

-0.60

shire

-0.60

POSITIVE LOGITS

tle

1.11

ter

0.97

ters

0.94

offs

0.93

chell

0.89

ups

0.89

list

0.85

lists

0.83

tering

0.83

meal

0.81

Activations Density 0.045%

specific mentions of the word "set."

references to a collection or group of items

No Comments

No Known Activations