INDEX

Explanations

mentions of specific names and titles, potentially within news articles or reports

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names related to significant events, people, or organizations

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

ĸļ

-0.65

EStream

-0.63

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.62

è¦ļéĨĴ

-0.61

 respectively

-0.59

 thereof

-0.57

Ïī

-0.55

Magikarp

-0.53

}.

-0.53

 precursor

-0.53

POSITIVE LOGITS

resa

0.73

ifax

0.68

ley

0.67

Berry

0.65

orm

0.63

ola

0.63

inking

0.62

lyn

0.62

anyahu

0.62

ston

0.62

Activations Density 0.580%

mentions of specific names and titles, potentially within news articles or reports

proper nouns or names related to significant events, people, or organizations

No Comments

No Known Activations