INDEX

Explanations

strong indications or hints

oai_token-act-pair · gpt-3.5-turbo

references to hints or evidence related to mysteries or investigations

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

kus

-0.73

imposed

-0.67

don

-0.67

ntil

-0.65

requency

-0.65

rior

-0.65

artney

-0.63

EVA

-0.62

eg

-0.61

tem

-0.60

POSITIVE LOGITS

 clue

1.41

 hint

1.10

 clues

1.01

wcs

0.86

 hints

0.79

lycer

0.72

hole

0.72

TextColor

0.71

 glean

0.70

llor

0.68

Activations Density 0.010%

strong indications or hints

references to hints or evidence related to mysteries or investigations

No Comments

No Known Activations