INDEX

Explanations

random letters and symbols that do not seem to form coherent words or phrases

oai_token-act-pair · gpt-3.5-turbo

non-standard or unusual formatting and encoding characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

gad

-0.81

anwhile

-0.77

wagen

-0.76

Peb

-0.75

 scattering

-0.73

nesday

-0.70

 Berm

-0.69

 dispers

-0.68

 messenger

-0.67

 wink

-0.67

POSITIVE LOGITS

º

1.34

¹

1.27

£

1.21

Į

1.16

®

1.13

į

1.13

¬

1.11

ı

1.09

ª

1.09

Ī

1.07

Activations Density 0.100%

random letters and symbols that do not seem to form coherent words or phrases

non-standard or unusual formatting and encoding characters

No Comments

No Known Activations