INDEX

Explanations

words related to topics or names containing "th"

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "th."

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

staking

-0.75

ITED

-0.75

 affairs

-0.66

 Canaver

-0.63

GMT

-0.63

-0.62

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.60

BIL

-0.60

 portfolios

-0.58

Rules

-0.58

POSITIVE LOGITS

ulhu

1.30

ttp

1.18

orne

1.13

umbnail

1.13

ousands

1.03

urst

1.01

ieves

0.96

ieving

0.96

istle

0.95

ousand

0.95

Activations Density 0.039%

words related to topics or names containing "th"

the occurrence of the word "th."

No Comments

No Known Activations