INDEX

Explanations

questions or prompts indicated by a question mark at the end of a sentence

oai_token-act-pair · gpt-3.5-turbo

questions and inquiries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cipled

-0.72

 fanc

-0.72

fut

-0.70

 tables

-0.66

 migr

-0.65

lap

-0.65

anyahu

-0.63

 transgress

-0.63

bidden

-0.63

opath

-0.63

POSITIVE LOGITS

 Seems

1.15

 Would

1.14

 Obviously

1.13

 Explain

1.11

 Answer

1.10

 Personally

1.06

 Especially

1.01

 Does

0.99

 Were

0.99

Did

0.98

Activations Density 0.083%

questions or prompts indicated by a question mark at the end of a sentence

questions and inquiries

No Comments

No Known Activations