INDEX

Explanations

authors and studies mentioned in research papers

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate authorship or references to reports and studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

pring

-0.70

llan

-0.65

 twitch

-0.63

Enlarge

-0.62

 dude

-0.62

hunt

-0.61

 vibration

-0.61

 broom

-0.61

RL

-0.60

_-

-0.59

POSITIVE LOGITS

 books

0.78

 articles

0.77

 Letters

0.74

letters

0.74

 bestselling

0.72

Ô

0.72

 novels

0.71

 memoir

0.71

 unpublished

0.70

 poems

0.69

Activations Density 0.078%

authors and studies mentioned in research papers

phrases that indicate authorship or references to reports and studies

No Comments

No Known Activations