INDEX

Explanations

items related to physical objects and modifications

oai_token-act-pair · gpt-3.5-turbo

punctuation and symbols, particularly commas and colons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

":"/

-0.90

¬¼

-0.80

Ĥİ

-0.78

²¾

-0.76

Ħ¢

-0.75

rely

-0.68

owe

-0.68

Ń·

-0.68

®

-0.65

ITE

-0.64

POSITIVE LOGITS

 plus

1.35

 which

1.04

 coupled

0.98

 followed

0.97

 namely

0.97

 along

0.93

 including

0.91

 plenty

0.90

etc

0.89

 accompanied

0.89

Activations Density 0.363%

items related to physical objects and modifications

punctuation and symbols, particularly commas and colons

No Comments

No Known Activations