INDEX

Explanations

prominent newspaper names

oai_token-act-pair · gpt-3.5-turbo

references to specific newspapers and media outlets

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

natureconservancy

-0.74

milo

-0.72

avorite

-0.72

afety

-0.69

inventoryQuantity

-0.68

worldly

-0.68

blance

-0.68

Rating

-0.67

 fixme

-0.67

 portfolios

-0.66

POSITIVE LOGITS

stice

0.71

 Skydragon

0.66

 Rays

0.66

 Baptist

0.65

 Yankees

0.62

 Angels

0.61

ados

0.61

 Tribune

0.59

iffs

0.59

 Lines

0.59

Activations Density 0.126%

prominent newspaper names

references to specific newspapers and media outlets

No Comments

No Known Activations