INDEX

Explanations

titles and organizations, particularly those related to law enforcement and government

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'L' in varied contexts

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

Negative Logits

 Canaan

-0.68

 Zoro

-0.68

 Bigfoot

-0.65

 Celest

-0.64

 Negro

-0.63

 comprom

-0.63

 Garmin

-0.62

 fortun

-0.61

 beaut

-0.61

 Gaul

-0.61

POSITIVE LOGITS

ï¸ı

0.92

vernment

0.91

tarian

0.90

TION

0.87

rative

0.85

intendent

0.83

SHIP

0.82

ulty

0.80

tre

0.80

ship

0.79

Activations Density 0.392%

titles and organizations, particularly those related to law enforcement and government

instances of the letter 'L' in varied contexts

No Known Activations