INDEX

Explanations

instances where something has been achieved or accomplished

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "doing it again" or similar expressions indicating repetition

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

afa

-0.61

yond

-0.61

 Proceed

-0.61

 contracted

-0.60

rish

-0.60

 obser

-0.60

ancel

-0.59

awa

-0.59

SPONSORED

-0.59

currently

-0.58

POSITIVE LOGITS

 Heller

0.75

 plenty

0.67

 brilliantly

0.63

 nightmare

0.60

ById

0.59

 Plenty

0.56

ãĤ¤ãĥĪ

0.56

izophren

0.56

 Wasserman

0.56

 awfully

0.55

Activations Density 0.558%

instances where something has been achieved or accomplished

instances of the phrase "doing it again" or similar expressions indicating repetition

No Comments

No Known Activations