INDEX

Explanations

proper nouns related to locations ending with "burg."

oai_token-act-pair · gpt-3.5-turbo

references to the word "burg" in various contexts

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

ISE

-0.71

::::::::

-0.68

Mex

-0.67

TAM

-0.67

wcsstore

-0.64

 Gemini

-0.63

MAL

-0.63

ãĥ¼ãĥ³

-0.61

ãĥŀ

-0.60

 iTunes

-0.60

POSITIVE LOGITS

burg

1.24

sburgh

1.16

undy

1.10

burgh

1.04

sburg

0.98

raltar

0.95

sylvania

0.95

lar

0.95

hester

0.91

culosis

0.91

Activations Density 0.008%

proper nouns related to locations ending with "burg."

references to the word "burg" in various contexts

No Comments

No Known Activations