INDEX

Explanations

questions or requests for information, typically involving specifics and details

oai_token-act-pair · gpt-3.5-turbo

questions related to costs, processes, and distributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

©¶æ

-0.70

 Corona

-0.66

 Kers

-0.65

 laure

-0.65

Ĺ

-0.59

frog

-0.58

Ö¼

-0.58

coat

-0.58

 Gree

-0.58

ffen

-0.57

POSITIVE LOGITS

how

0.94

 whether

0.87

whether

0.87

HOW

0.86

how

0.86

thumbnails

0.82

 whereabouts

0.81

why

0.78

HOW

0.78

WHY

0.75

Activations Density 0.399%

questions or requests for information, typically involving specifics and details

questions related to costs, processes, and distributions

No Comments

No Known Activations