INDEX

Explanations

phrases related to providing explanations, examples, or arguments

oai_token-act-pair · gpt-3.5-turbo

the word "which" in various contexts

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

dj

-0.79

rolet

-0.77

ovo

-0.73

tty

-0.73

ondon

-0.73

soType

-0.73

bath

-0.71

apult

-0.70

rene

-0.70

redit

-0.69

POSITIVE LOGITS

upon

1.20

soever

1.11

he

0.93

 they

0.92

 case

0.88

she

0.85

 contestants

0.81

we

0.76

 viewers

0.75

 cases

0.74

Activations Density 0.049%

phrases related to providing explanations, examples, or arguments

the word "which" in various contexts

No Comments

No Known Activations