INDEX

Explanations

phrases related to items or concepts that are named or referred to with the word "back"

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "back" 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ISION

-0.68

IFIED

-0.65

vez

-0.63

 ILCS

-0.63

 Parenthood

-0.63

itta

-0.62

urious

-0.61

 Aires

-0.60

oppable

-0.60

Irish

-0.59

POSITIVE LOGITS

wards

1.12

door

0.93

lash

0.89

packing

0.79

doors

0.77

drops

0.77

gam

0.76

vironment

0.74

ward

0.74

GROUND

0.72

Activations Density 0.031%

phrases related to items or concepts that are named or referred to with the word "back"

references to the concept of "back" in various contexts

No Comments

No Known Activations