INDEX

Explanations

phrases indicating causation or influence

oai_token-act-pair · gpt-3.5-turbo

phrases related to cause and effect relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ntax

-0.65

 Else

-0.64

amily

-0.62

ntil

-0.62

 Tacoma

-0.62

 Croat

-0.62

ilion

-0.60

Kal

-0.60

utenberg

-0.60

 Challenger

-0.60

POSITIVE LOGITS

 attracts

0.75

 undermines

0.74

 prompts

0.74

SPONSORED

0.73

 generates

0.72

 inspires

0.72

 creates

0.72

 contributes

0.71

 embold

0.70

 leads

0.69

Activations Density 0.041%

phrases indicating causation or influence

phrases related to cause and effect relationships

No Comments

No Known Activations