INDEX

Explanations

phrases related to providing guidance, instructions, or information

oai_token-act-pair · gpt-3.5-turbo

terms related to instructional content or guides

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Corpus

-0.73

itol

-0.72

inition

-0.70

 move

-0.70

 Scarborough

-0.66

abase

-0.66

 measure

-0.65

Baltimore

-0.65

ynthesis

-0.65

 dataset

-0.64

POSITIVE LOGITS

 guides

3.08

 Guides

2.20

 interpre

1.20

 hacks

1.13

 helpers

1.13

 instructors

1.10

 tutorials

1.10

 mounts

1.08

 anchors

1.05

 seals

1.00

Activations Density 0.019%

phrases related to providing guidance, instructions, or information

terms related to instructional content or guides

No Comments

No Known Activations