INDEX

Explanations

proper nouns related to various locations, people, and organizations

oai_token-act-pair · gpt-3.5-turbo

phrases related to critical assessments or reviews

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

-0.68

lass

-0.66

']

-0.65

 Recomm

-0.64

.?

-0.64

'/

-0.64

lette

-0.64

without

-0.64

"],"

-0.64

STEM

-0.63

POSITIVE LOGITS

 exception

0.91

 exceptions

0.90

 emphasis

0.85

 caveat

0.79

 emph

0.73

 notable

0.73

 caveats

0.71

 thrown

0.70

 twist

0.70

 hindsight

0.69

Activations Density 0.427%

proper nouns related to various locations, people, and organizations

phrases related to critical assessments or reviews

No Comments

No Known Activations