INDEX

Explanations

expressions of dissatisfaction or disapproval

oai_token-act-pair · gpt-3.5-turbo

phrases expressing negative evaluations or dissatisfaction

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

beit

-0.70

lessness

-0.69

azar

-0.68

ription

-0.66

onomy

-0.64

osen

-0.61

rather

-0.61

empt

-0.60

uggest

-0.59

ump

-0.59

POSITIVE LOGITS

 anymore

1.48

nor

1.14

 either

0.82

ãĤ®

0.76

yet

0.76

Ú

0.75

yet

0.74

 anywhere

0.69

 necessarily

0.69

 Enough

0.69

Activations Density 0.314%

expressions of dissatisfaction or disapproval

phrases expressing negative evaluations or dissatisfaction

No Comments

No Known Activations