INDEX

Explanations

terms related to historical landmarks or structures

oai_token-act-pair · gpt-3.5-turbo

terms related to cultural and geographical identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.62

GD

-0.60

OLOGY

-0.59

boat

-0.58

 decom

-0.58

sheet

-0.57

 overlay

-0.57

 correction

-0.56

 Robbins

-0.56

 strings

-0.55

POSITIVE LOGITS

hire

1.14

cence

1.13

cript

1.12

aturated

1.03

cent

1.01

kaya

0.98

ilver

0.98

omes

0.96

erker

0.96

ystem

0.95

Activations Density 0.078%

terms related to historical landmarks or structures

terms related to cultural and geographical identities

No Comments

No Known Activations