INDEX

Explanations

instances of the word "admitting" and variations of the word "permitting"

oai_token-act-pair · gpt-3.5-turbo

terms related to permission or submission processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Greenberg

-0.68

 Woody

-0.67

 synergy

-0.65

pal

-0.63

 Swift

-0.63

tug

-0.61

 Sheen

-0.60

 Ronald

-0.59

 knock

-0.58

 Star

-0.58

POSITIVE LOGITS

mits

4.74

mitting

2.68

mit

2.13

missions

1.91

mitted

1.77

mitt

1.58

MIT

1.48

mission

1.29

mitter

1.27

lees

1.06

Activations Density 0.013%

instances of the word "admitting" and variations of the word "permitting"

terms related to permission or submission processes

No Comments

No Known Activations