INDEX

Explanations

comparative statements expressing a higher magnitude or level

oai_token-act-pair · gpt-3.5-turbo

expressions emphasizing the concept of 'more' or comparisons involving additional information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

jam

-0.71

jab

-0.70

ipel

-0.69

escal

-0.68

oes

-0.67

whatever

-0.67

ement

-0.66

inished

-0.64

alysis

-0.64

uum

-0.63

POSITIVE LOGITS

 than

0.80

 importantly

0.78

 besides

0.78

HUD

0.75

ÂłÂłÂłÂł

0.72

SPONSORED

0.71

 astounding

0.66

 informative

0.64

 noteworthy

0.64

NB

0.63

Activations Density 0.046%

comparative statements expressing a higher magnitude or level

expressions emphasizing the concept of 'more' or comparisons involving additional information

No Comments

No Known Activations