INDEX

Explanations

the word "Tu" followed by a number

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "Tu"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

éĹĺ

-0.76

å°Ĩ

-0.75

Ö¼

-0.73

manship

-0.72

âķĲ

-0.72

âķĲâķĲ

-0.70

âĸ¬

-0.67

 Rebirth

-0.67

SPD

-0.67

é¾į

-0.67

POSITIVE LOGITS

fts

1.09

Tu

1.07

ples

1.03

culosis

1.00

Tu

0.93

esta

0.87

zan

0.84

 Klux

0.83

tu

0.83

zin

0.82

Activations Density 0.009%

the word "Tu" followed by a number

repeated references to the word "Tu"

No Comments

No Known Activations