INDEX

Explanations

the word 'and'

oai_token-act-pair · gpt-3.5-turbo

conjunctions that connect ideas or phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

wic

-0.68

 unfocusedRange

-0.66

witz

-0.66

andom

-0.65

resy

-0.64

osit

-0.63

writ

-0.62

ieval

-0.62

eral

-0.61

 eternity

-0.61

POSITIVE LOGITS

uras

0.69

 migraine

0.65

 Antar

0.63

seekers

0.63

interstitial

0.62

ntil

0.61

 bulbs

0.61

 seat

0.60

 Higgins

0.60

inus

0.59

Activations Density 0.000%

the word 'and'

conjunctions that connect ideas or phrases

No Comments

No Known Activations