INDEX

Explanations

the total score or final result of a competition or evaluation

oai_token-act-pair · gpt-3.5-turbo

references to total scores or aggregate values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 visibility

-0.69

 advis

-0.65

 orientation

-0.64

aga

-0.63

continue

-0.62

 explanations

-0.62

lean

-0.61

nav

-0.61

vag

-0.60

clus

-0.60

POSITIVE LOGITS

OTAL

0.76

 Monstrous

0.70

iaries

0.69

 GDDR

0.67

 Dresden

0.65

vertisement

0.65

Constructed

0.64

 Fires

0.64

ãģ®éŃĶ

0.64

 Thrones

0.63

Activations Density 0.000%

the total score or final result of a competition or evaluation

references to total scores or aggregate values

No Comments

No Known Activations