INDEX

Explanations

dates, particularly years, presented in a specific format

oai_token-act-pair · gpt-3.5-turbo

specific years and their corresponding months

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

yip

-0.78

ibilities

-0.63

eno

-0.61

Hue

-0.59

ss

-0.58

 Salam

-0.58

pires

-0.55

hr

-0.55

etz

-0.54

eds

-0.54

POSITIVE LOGITS

uilt

0.67

VK

0.66

ocent

0.65

Clean

0.64

kel

0.63

pez

0.63

oval

0.63

å¤©

0.62

ggy

0.62

 livest

0.61

Activations Density 0.079%

dates, particularly years, presented in a specific format

specific years and their corresponding months

No Comments

No Known Activations