INDEX

Explanations

phrases related to seeking or providing information

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate uncertainty or hedging in statements

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

ãĤ©

-0.82

âĶĢâĶĢâĶĢâĶĢ

-0.82

quit

-0.78

QB

-0.77

BUT

-0.73

ãĥĺ

-0.73

olution

-0.71

utic

-0.70

ourney

-0.70

à¨

-0.70

POSITIVE LOGITS

 anecd

1.36

 suffice

1.13

 anecdotal

0.95

 indications

0.91

 sources

0.89

 rumors

0.88

 statements

0.86

 analysts

0.86

 insiders

0.85

 excerpts

0.84

Activations Density 0.161%

phrases related to seeking or providing information

phrases that indicate uncertainty or hedging in statements

No Comments

No Known Activations