INDEX

Explanations

phrases indicating conditions or requirements before a specific action is taken

oai_token-act-pair · gpt-3.5-turbo

the word "only" in various contexts, often indicating limitations or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

acca

-0.75

aston

-0.75

ongyang

-0.74

Mods

-0.73

WT

-0.73

arez

-0.72

ott

-0.71

redit

-0.68

ãĤ¦ãĤ¹

-0.68

UD

-0.68

POSITIVE LOGITS

etheless

1.07

 nonetheless

1.05

 nevertheless

0.82

 overshadowed

0.81

 importantly

0.80

 retains

0.79

 retained

0.73

 succeeded

0.73

 persists

0.72

 remains

0.71

Activations Density 0.135%

phrases indicating conditions or requirements before a specific action is taken

the word "only" in various contexts, often indicating limitations or conditions

No Comments

No Known Activations