INDEX

Explanations

terms related to creating user-friendly experiences

oai_token-act-pair · gpt-3.5-turbo

terms related to user-friendliness and inclusivity in design

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 buried

-0.69

aman

-0.63

pec

-0.63

Heart

-0.63

Bone

-0.62

kar

-0.60

PIN

-0.60

ulet

-0.60

atro

-0.58

xxxxxxxx

-0.58

POSITIVE LOGITS

lihood

0.73

glers

0.73

recy

0.68

tenance

0.66

friendly

0.63

 skew

0.62

 environments

0.61

 underdog

0.60

minist

0.60

 behaviors

0.59

Activations Density 0.066%

terms related to creating user-friendly experiences

terms related to user-friendliness and inclusivity in design

No Comments

No Known Activations