INDEX

Explanations

similarities or comparisons between different items or concepts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "similar" and related comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

OST

-0.79

olate

-0.71

ole

-0.67

arden

-0.67

Ã«

-0.66

oway

-0.65

Wah

-0.65

 Loaded

-0.65

arer

-0.63

don

-0.63

POSITIVE LOGITS

lihood

1.05

worldly

1.02

 minded

0.96

quartered

0.93

 vein

0.91

minded

0.90

 amounts

0.90

 analogous

0.89

MpServer

0.86

etheless

0.85

Activations Density 0.020%

similarities or comparisons between different items or concepts

instances of the word "similar" and related comparisons

No Comments

No Known Activations