INDEX

Explanations

connected phrases or pairs of items

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases indicating relationships or connections

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

ider

-0.71

dayName

-0.65

icc

-0.65

izer

-0.63

enium

-0.63

iste

-0.63

SPA

-0.63

qt

-0.61

iets

-0.61

ct

-0.61

POSITIVE LOGITS

 secondly

1.06

 furthermore

1.04

 consequently

0.87

 finally

0.87

theless

0.80

 THEN

0.78

 therefore

0.77

 moreover

0.76

 importantly

0.76

 thence

0.76

Activations Density 0.066%

connected phrases or pairs of items

conjunctions and phrases indicating relationships or connections

No Comments

No Known Activations