INDEX

Explanations

combinations of elements or factors

oai_token-act-pair · gpt-3.5-turbo

phrases that describe combinations of elements or factors

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

artisan

-0.70

yers

-0.70

shr

-0.68

cients

-0.67

ached

-0.67

arers

-0.67

ternity

-0.66

ond

-0.66

ser

-0.65

ÅĤ

-0.65

POSITIVE LOGITS

 combination

0.98

 combinations

0.96

ãĤ«

0.79

ãĤ¤ãĥĪ

0.76

ocation

0.74

 thereof

0.74

ãĤ¼ãĤ¦ãĤ¹

0.72

uations

0.71

 mixture

0.70

ãĥ¼ãĥ³

0.70

Activations Density 0.012%

combinations of elements or factors

phrases that describe combinations of elements or factors

No Comments

No Known Activations