INDEX

Explanations

non-standard characters and symbols at the end of words

oai_token-act-pair · gpt-3.5-turbo

negative contractions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

RAD

-0.76

çīĪ

-0.68

 guiActiveUnfocused

-0.66

 Samar

-0.62

 scapego

-0.61

 mosaic

-0.60

 camer

-0.60

 Shib

-0.60

 guiActiveUn

-0.60

jug

-0.59

POSITIVE LOGITS

£

1.15

¹

0.99

º

0.95

¿

0.95

¼

0.93

Ń

0.91

İ

0.91

¬

0.91

Ķ

0.90

ĵ

0.90

Activations Density 0.148%

non-standard characters and symbols at the end of words

negative contractions

No Comments

No Known Activations