INDEX

Explanations

personal pronouns indicating possession or reference

oai_token-act-pair · gpt-3.5-turbo

pronouns that refer to individuals and groups

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

 Course

-0.66

 srfAttach

-0.66

ories

-0.66

 Sloan

-0.65

 Goldstein

-0.61

 Coalition

-0.61

 Beckham

-0.61

 dawn

-0.61

gee

-0.61

cemic

-0.61

POSITIVE LOGITS

 hooked

0.83

atic

0.76

 reconsider

0.75

 deported

0.74

 laugh

0.71

 addicted

0.69

 fooled

0.69

 pegged

0.68

 hanged

0.68

 extrad

0.68

Activations Density 0.109%

personal pronouns indicating possession or reference

pronouns that refer to individuals and groups

No Comments

No Known Activations