INDEX

Explanations

instances of personal prompting or questioning

oai_token-act-pair · gpt-3.5-turbo

interrogative and conversational phrases directed at the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

çīĪ

-0.67

Austral

-0.63

»Ĵ

-0.61

Net

-0.60

 Pathfinder

-0.60

Marginal

-0.59

gypt

-0.58

Magikarp

-0.58

 Austral

-0.57

 Mirage

-0.57

POSITIVE LOGITS

iety

0.80

 anymore

0.79

yet

0.79

vable

0.75

hett

0.71

existent

0.70

iable

0.69

bered

0.69

âĻ

0.68

bley

0.67

Activations Density 0.161%

instances of personal prompting or questioning

interrogative and conversational phrases directed at the reader

No Comments

No Known Activations