INDEX

Explanations

references to written or spoken communication

oai_token-act-pair · gpt-3.5-turbo

significant actions or events that are likely to have an impact or provoke a reaction

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Quan

-0.67

 transmissions

-0.63

 Fairfax

-0.61

 Stars

-0.59

 divergence

-0.58

ACS

-0.57

 Ware

-0.57

 Engineers

-0.57

 Communications

-0.56

TTL

-0.56

POSITIVE LOGITS

 dearly

0.89

 sorely

0.74

 overlook

0.73

 deem

0.73

assy

0.68

GB

0.67

 emulate

0.67

ãĥ¼ãĥĨãĤ£

0.66

ãĤ´ãĥ³

0.66

alys

0.65

Activations Density 0.327%

references to written or spoken communication

significant actions or events that are likely to have an impact or provoke a reaction

No Comments

No Known Activations