INDEX

Explanations

phrases indicating future predictions or intentions

oai_token-act-pair · gpt-3.5-turbo

references to increasing amounts or quantities of something

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

escription

-0.74

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.61

 Cipher

-0.61

iasis

-0.60

;;;;;;;;;;;;

-0.60

«ĺ

-0.60

¬¼

-0.59

 Garrison

-0.59

 Bohem

-0.58

 Sheen

-0.57

POSITIVE LOGITS

 than

1.69

than

1.60

 Than

1.20

efficient

0.80

rounded

0.70

rab

0.69

catentry

0.66

 besides

0.66

rah

0.65

erous

0.65

Activations Density 0.444%

phrases indicating future predictions or intentions

references to increasing amounts or quantities of something

No Comments

No Known Activations