INDEX

Explanations

phrases related to legal and business contexts, specifically involving patents and competitors

oai_token-act-pair · gpt-3.5-turbo

references to a significant entity or concept in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

leans

-0.74

©¶æ

-0.70

 prohibited

-0.66

 peeled

-0.65

 subscribed

-0.61

 Empire

-0.61

 obligated

-0.60

oug

-0.60

 stricken

-0.60

 sucked

-0.59

POSITIVE LOGITS

agents

0.89

Thumbnail

0.83

alter

0.83

ampions

0.82

avez

0.80

aux

0.78

Mit

0.74

ampion

0.73

cone

0.71

ulence

0.71

Activations Density 0.000%

phrases related to legal and business contexts, specifically involving patents and competitors

references to a significant entity or concept in context

No Comments

No Known Activations