INDEX

Explanations

terms related to the word "nec" or "necrosis"

oai_token-act-pair · gpt-3.5-turbo

terms related to necessity or essential concepts

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

 Feld

-0.85

 Wink

-0.75

xual

-0.73

 Kafka

-0.68

zac

-0.67

boards

-0.66

âķĲâķĲ

-0.65

 Sawyer

-0.65

 Beckham

-0.65

nings

-0.65

POSITIVE LOGITS

romancer

1.24

ropolis

1.03

hemy

0.92

Nec

0.91

nec

0.89

nec

0.88

 cumbers

0.87

ensibly

0.85

ategory

0.85

emic

0.85

Activations Density 0.011%

terms related to the word "nec" or "necrosis"

terms related to necessity or essential concepts

No Comments

No Known Activations