INDEX

Explanations

locations, specifically those related to Australia

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "au"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

STATE

-0.64

ACTED

-0.64

 flares

-0.61

Wallet

-0.60

selves

-0.60

 Cavaliers

-0.59

shows

-0.59

APS

-0.59

bread

-0.58

 GOODMAN

-0.58

POSITIVE LOGITS

llah

1.13

gment

1.04

qua

1.00

lette

0.98

ction

0.97

fman

0.93

vre

0.87

cel

0.85

cham

0.84

pload

0.83

Activations Density 0.020%

locations, specifically those related to Australia

mentions of the term "au"

No Comments

No Known Activations