INDEX

Explanations

phrases related to comparison or contrast

oai_token-act-pair · gpt-3.5-turbo

references to the word "well."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anos

-0.78

pid

-0.65

anon

-0.64

mare

-0.62

ierce

-0.62

anus

-0.61

zan

-0.61

absolute

-0.61

mid

-0.60

agara

-0.60

POSITIVE LOGITS

 evidenced

0.61

 optionally

0.60

NESS

0.60

onse

0.59

umenthal

0.58

FTWARE

0.58

 insofar

0.57

Label

0.57

 vers

0.57

 possibly

0.56

Activations Density 0.030%

phrases related to comparison or contrast

references to the word "well."

No Comments

No Known Activations