INDEX

Explanations

phrases involving the location "there."

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "there" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 waiver

-0.61

 Stras

-0.59

Dev

-0.58

 resil

-0.58

 Nanto

-0.58

 Expend

-0.58

iencies

-0.57

 Warfare

-0.56

 Fallen

-0.56

eer

-0.55

POSITIVE LOGITS

abouts

2.01

upon

1.34

tics

1.18

tical

1.18

with

1.09

from

1.05

fore

1.02

for

0.95

tic

0.92

comings

0.89

Activations Density 0.061%

phrases involving the location "there."

mentions of the word "there" in various contexts

No Comments

No Known Activations