INDEX

Explanations

phrases involving the concept of going or being backwards

oai_token-act-pair · gpt-3.5-turbo

references to performing actions in reverse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

rament

-0.83

raltar

-0.80

ateurs

-0.79

"},"

-0.75

chens

-0.75

riz

-0.75

akings

-0.74

atum

-0.74

ulet

-0.73

anooga

-0.72

POSITIVE LOGITS

stairs

0.94

wards

0.94

ward

0.86

 compatibility

0.80

 compat

0.78

step

0.76

 spiral

0.73

WARD

0.72

fitted

0.71

side

0.70

Activations Density 0.017%

phrases involving the concept of going or being backwards

references to performing actions in reverse

No Comments

No Known Activations