INDEX

Explanations

dates or events in reported news articles

oai_token-act-pair · gpt-3.5-turbo

punctuation and contextual markers within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 pound

-0.69

 ethn

-0.66

 inward

-0.62

iod

-0.61

 individually

-0.60

 ambassadors

-0.59

uary

-0.59

 religiously

-0.58

edom

-0.58

 unfamiliar

-0.57

POSITIVE LOGITS

 Logged

1.27

SEE

1.27

Reviewer

1.19

 Photo

1.11

 Posted

1.05

 Loading

0.87

 View

0.87

 Listen

0.87

ļéĨĴ

0.85

<|endoftext|>

0.84

Activations Density 0.144%

dates or events in reported news articles

punctuation and contextual markers within the text

No Comments

No Known Activations