INDEX

Explanations

phrases emphasizing an opinion or statement/result

oai_token-act-pair · gpt-3.5-turbo

usage of punctuation marks and sentence endings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ible

-0.78

oid

-0.74

ilyn

-0.73

AGES

-0.73

rikes

-0.71

ovi

-0.71

ottesville

-0.69

oids

-0.68

ouver

-0.68

ened

-0.68

POSITIVE LOGITS

cipled

0.75

cffff

0.70

!--

0.69

âĢ¢âĢ¢âĢ¢âĢ¢

0.66

mington

0.65

 Zoro

0.65

onga

0.63

 sovere

0.63

Page

0.63

Ã°

0.63

Activations Density 0.010%

phrases emphasizing an opinion or statement/result

usage of punctuation marks and sentence endings

No Comments

No Known Activations