INDEX

Explanations

items associated with descriptions of objects or attributes

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate value or importance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

projects

-0.80

sports

-0.76

ouble

-0.75

GROUP

-0.74

Nation

-0.74

owship

-0.72

ateurs

-0.72

Train

-0.70

 Volunte

-0.70

onse

-0.70

POSITIVE LOGITS

 resemblance

1.13

 twist

1.04

 expiration

1.01

 characteristics

1.00

 diameter

0.98

 dimensions

0.97

 similarity

0.97

 inscription

0.96

 suffix

0.91

 purpose

0.91

Activations Density 0.370%

items associated with descriptions of objects or attributes

phrases that indicate value or importance

No Comments

No Known Activations