INDEX

Explanations

challenges or instances where challenges are mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to various challenges

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

otide

-0.80

tarian

-0.73

umn

-0.66

izations

-0.66

erie

-0.64

ahon

-0.63

ovies

-0.61

olia

-0.61

riter

-0.61

ool

-0.61

POSITIVE LOGITS

 posed

1.13

 faced

0.84

enges

0.81

 facing

0.76

 insur

0.76

olving

0.76

 daunting

0.75

 confronting

0.75

 challenge

0.75

enged

0.75

Activations Density 0.059%

challenges or instances where challenges are mentioned in a text

references to various challenges

No Comments

No Known Activations