INDEX

Explanations

descriptions indicating how something is mentioned or described

oai_token-act-pair · gpt-3.5-turbo

phrases that simplify or clarify complex ideas

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Skydragon

-0.71

incible

-0.65

IELD

-0.64

imura

-0.63

ds

-0.63

Legendary

-0.63

hung

-0.62

owners

-0.59

wives

-0.59

gered

-0.59

POSITIVE LOGITS

 bluntly

1.39

 mildly

1.21

 succinct

1.20

 plainly

1.15

 politely

1.12

 nicely

1.03

 concise

0.99

 gently

0.91

 diplom

0.90

 lightly

0.89

Activations Density 0.070%

descriptions indicating how something is mentioned or described

phrases that simplify or clarify complex ideas

No Comments

No Known Activations