INDEX

Explanations

phrases indicating conversion or transformation from one form to another

oai_token-act-pair · gpt-3.5-turbo

the phrase "to" followed by various nouns or actions, indicating transformations or changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

icious

-0.74

enture

-0.71

PLIED

-0.70

aturdays

-0.68

UI

-0.63

iltr

-0.62

autions

-0.62

Educ

-0.62

Wr

-0.61

 shenanigans

-0.61

POSITIVE LOGITS

 ç¥ŀ

0.73

©¶æ

0.69

 reflect

0.63

kus

0.63

 Catholicism

0.62

 Sonny

0.62

livion

0.61

 accommodate

0.61

 resemble

0.61

 theirs

0.60

Activations Density 0.180%

phrases indicating conversion or transformation from one form to another

the phrase "to" followed by various nouns or actions, indicating transformations or changes

No Comments

No Known Activations