INDEX

Explanations

phrases indicating contrast or conditions

oai_token-act-pair · gpt-3.5-turbo

the word "not" and its context in relation to negation or contradiction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lance

-0.70

Shares

-0.70

plates

-0.70

iband

-0.67

kamp

-0.66

ription

-0.66

alez

-0.65

ngth

-0.64

vant

-0.64

riber

-0.64

POSITIVE LOGITS

 outright

1.12

 necessarily

0.99

 downright

0.95

ifies

0.77

 mistaken

0.75

icably

0.71

 intimidated

0.71

 exactly

0.71

 explicitly

0.71

 altogether

0.69

Activations Density 0.035%

phrases indicating contrast or conditions

the word "not" and its context in relation to negation or contradiction

No Comments

No Known Activations