INDEX

Explanations

language related to expressing opinion or making a decision

oai_token-act-pair · gpt-3.5-turbo

expressions indicating confidence or certainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Koran

-0.57

 Kissinger

-0.57

 Poverty

-0.56

cible

-0.56

 Kosovo

-0.56

ynamic

-0.53

bestos

-0.53

bris

-0.53

 Kabul

-0.52

DERR

-0.51

POSITIVE LOGITS

.ï¿½

1.11

 ðŁĺ

0.96

.ãĢį

0.96

 Â¯

0.93

 ðŁĻĤ

0.93

âĢ

0.90

.<

0.88

âķĲâķĲ

0.86

soType

0.86

.</

0.86

Activations Density 0.685%

language related to expressing opinion or making a decision

expressions indicating confidence or certainty

No Comments

No Known Activations