INDEX

Explanations

phrases indicating confusion or uncertainty

oai_token-act-pair · gpt-3.5-turbo

references to a lack of knowledge or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

iki

-0.74

inka

-0.71

Reviewed

-0.68

ouri

-0.67

inos

-0.66

Pers

-0.64

istant

-0.63

ansk

-0.63

conn

-0.62

visor

-0.60

POSITIVE LOGITS

 whatsoever

0.98

how

0.97

why

0.95

 whats

0.82

 what

0.80

 whence

0.80

ledged

0.79

why

0.77

 squat

0.75

WHY

0.74

Activations Density 0.036%

phrases indicating confusion or uncertainty

references to a lack of knowledge or uncertainty

No Comments

No Known Activations