INDEX

Explanations

words related to comparison or similarity

oai_token-act-pair · gpt-3.5-turbo

phrases indicating comparisons or similarities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Bild

-0.81

 Derby

-0.72

ender

-0.69

 Mush

-0.68

 Ãĸ

-0.67

 Ventura

-0.67

 Stockholm

-0.67

 Oslo

-0.64

oster

-0.63

 Dota

-0.63

POSITIVE LOGITS

 minded

0.86

etheless

0.85

æ©Ł

0.84

 situated

0.82

ctr

0.82

quartered

0.81

isSpecialOrderable

0.79

 strugg

0.74

DEF

0.74

theless

0.72

Activations Density 0.011%

words related to comparison or similarity

phrases indicating comparisons or similarities

No Comments

No Known Activations