INDEX

Explanations

proper nouns representing people, places, or organizations

oai_token-act-pair · gpt-3.5-turbo

references to singular nouns or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 creeps

-0.76

 corpses

-0.75

anders

-0.74

enery

-0.73

 mutants

-0.73

 Koreans

-0.72

ppings

-0.71

ources

-0.71

ians

-0.70

verb

-0.70

POSITIVE LOGITS

 prestigious

0.86

 conversation

0.82

 survey

0.81

ĸļ

0.78

 conference

0.78

 weekly

0.76

 discussion

0.76

 consortium

0.74

 comprehensive

0.74

 monthly

0.74

Activations Density 0.533%

proper nouns representing people, places, or organizations

references to singular nouns or subjects

No Comments

No Known Activations