INDEX

Explanations

proper nouns or names preceded by "PT"

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific organization or entity, identified by the token "PT"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

hold

-0.95

minist

-0.91

cle

-0.75

combe

-0.71

cles

-0.69

gie

-0.69

bye

-0.68

lihood

-0.67

flies

-0.67

rers

-0.66

POSITIVE LOGITS

EMBER

0.93

OGR

0.91

BN

0.90

ypes

0.90

ION

0.90

OTO

0.88

BF

0.86

 Cruiser

0.83

BD

0.80

OLOG

0.80

Activations Density 0.018%

proper nouns or names preceded by "PT"

mentions of a specific organization or entity, identified by the token "PT"

No Comments

No Known Activations