INDEX

Explanations

phrases related to a specific person named Herzog

oai_token-act-pair · gpt-3.5-turbo

references to the figure "Herzog" and related concepts

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

Reviewer

-0.84

-0.72

ãĥ´ãĤ¡

-0.70

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.69

Spread

-0.67

ITAL

-0.63

âĶĢâĶĢâĶĢâĶĢ

-0.62

shapeshifter

-0.61

âķĲ

-0.61

CEPT

-0.60

POSITIVE LOGITS

 Herz

1.43

etz

0.89

icken

0.87

feld

0.84

icol

0.84

azard

0.83

uve

0.82

og

0.81

naires

0.81

anan

0.78

Activations Density 0.009%

phrases related to a specific person named Herzog

references to the figure "Herzog" and related concepts

No Comments

No Known Activations