INDEX

Explanations

special characters and formatting symbols

oai_token-act-pair · gpt-3.5-turbo

repeated asterisks or bullet points

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

rolet

-0.80

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.77

 migr

-0.73

 wisdom

-0.67

²¾

-0.67

 migration

-0.67

 fireplace

-0.67

vag

-0.66

 facing

-0.66

 grazing

-0.65

POSITIVE LOGITS

âĢ¢âĢ¢

0.95

âĢł

0.92

.*

0.90

âĨĳ

0.90

Bah

0.84

(*

0.83

AUT

0.83

Thompson

0.82

TON

0.81

--------------------------------------------------------

0.81

Activations Density 0.016%

special characters and formatting symbols

repeated asterisks or bullet points

No Comments

No Known Activations