INDEX

Explanations

words related to directions or pathways, particularly indicating movement or location

oai_token-act-pair · gpt-3.5-turbo

terms related to traffic direction and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

MAT

-0.88

soDeliveryDate

-0.79

ptive

-0.73

ATT

-0.70

otrop

-0.68

EAR

-0.68

ãģŁ

-0.68

ORY

-0.68

PROV

-0.68

MED

-0.66

POSITIVE LOGITS

bound

0.97

 lanes

0.91

naire

0.83

uction

0.82

interstitial

0.80

loads

0.80

duty

0.78

pread

0.76

 volumes

0.73

essel

0.73

Activations Density 0.005%

words related to directions or pathways, particularly indicating movement or location

terms related to traffic direction and conditions

No Comments

No Known Activations