INDEX

Explanations

occurrences of the word "wards" or the word "forwards."

oai_token-act-pair · gpt-3.5-turbo

instances of directional words related to movement

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

anan

-0.84

Luck

-0.67

 Mald

-0.62

zan

-0.61

nce

-0.61

Bird

-0.61

hoff

-0.60

WN

-0.60

Chain

-0.60

MENT

-0.60

POSITIVE LOGITS

wards

1.27

creen

1.14

peak

0.85

pring

0.81

hift

0.80

ileaks

0.79

jriwal

0.78

ervative

0.77

hops

0.77

uits

0.77

Activations Density 0.012%

occurrences of the word "wards" or the word "forwards."

instances of directional words related to movement

No Comments

No Known Activations