INDEX

Explanations

phrases expressing commands or instructions

oai_token-act-pair · gpt-3.5-turbo

negative contractions that express advice or caution

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥķãĤ©

-0.75

ItemThumbnailImage

-0.70

 permissible

-0.68

ilogy

-0.66

 logger

-0.60

 mosqu

-0.60

MpServer

-0.60

channelAvailability

-0.60

 slug

-0.59

 clitor

-0.59

POSITIVE LOGITS

acy

0.78

oso

0.77

hips

0.76

cha

0.75

aton

0.75

lees

0.74

ates

0.72

otine

0.71

nant

0.71

ude

0.71

Activations Density 0.124%

phrases expressing commands or instructions

negative contractions that express advice or caution

No Comments

No Known Activations