INDEX

Explanations

definitions or mentions of dictionaries

oai_token-act-pair · gpt-3.5-turbo

references to dictionaries and encyclopedias

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

roach

-0.75

eem

-0.72

sent

-0.71

aints

-0.69

raising

-0.68

ded

-0.68

rd

-0.68

rity

-0.68

achine

-0.68

Wil

-0.66

POSITIVE LOGITS

 dictionary

1.23

 Dictionary

1.16

ãĥ¼ãĥĨãĤ£

0.98

ictionary

0.90

 translator

0.86

ãĥķãĤ¡

0.83

ãĥĥãĤ¯

0.83

tymology

0.80

abulary

0.78

ãĤ¢ãĥ«

0.77

Activations Density 0.007%

definitions or mentions of dictionaries

references to dictionaries and encyclopedias

No Comments

No Known Activations