INDEX

Explanations

phrases indicating a strong negative sentiment or denial

oai_token-act-pair · gpt-3.5-turbo

negations or statements indicating the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

rn

-0.73

Ott

-0.69

ean

-0.68

 Appears

-0.67

lish

-0.67

eals

-0.66

lean

-0.65

otton

-0.64

leaning

-0.64

staking

-0.64

POSITIVE LOGITS

 shortage

1.13

 doubt

1.12

 indication

1.05

 reason

0.98

 guarantee

0.96

 discern

0.95

 conceivable

0.92

xious

0.91

 longer

0.91

 denying

0.90

Activations Density 0.061%

phrases indicating a strong negative sentiment or denial

negations or statements indicating the absence of something

No Comments

No Known Activations