INDEX

Explanations

specific notes or messages within a text

oai_token-act-pair · gpt-3.5-turbo

formatted notes or annotations related to various topics

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

aced

-0.69

luaj

-0.66

athered

-0.63

icum

-0.62

 acting

-0.61

namese

-0.60

 wreck

-0.59

inventoryQuantity

-0.59

riter

-0.58

 pissed

-0.57

POSITIVE LOGITS

books

1.14

book

1.11

1.03

ably

1.01

BOOK

0.98

!:

0.88

:-

0.87

 Regarding

0.82

0.80

:,

0.80

Activations Density 0.021%

specific notes or messages within a text

formatted notes or annotations related to various topics

No Comments

No Known Activations