INDEX

Explanations

phrases indicating reasoning or explanation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate reasons or justifications

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

yss

-0.85

 ILCS

-0.79

chn

-0.77

bats

-0.75

OLOGY

-0.73

chin

-0.73

fman

-0.71

ps

-0.70

hess

-0.70

abytes

-0.70

POSITIVE LOGITS

 exist

0.78

 inaction

0.78

 doubt

0.75

 hesitation

0.75

 discrepancy

0.75

 dispute

0.74

why

0.74

 existence

0.72

 justify

0.72

 skepticism

0.72

Activations Density 0.099%

phrases indicating reasoning or explanation

phrases that indicate reasons or justifications

No Comments

No Known Activations