INDEX

Explanations

phrases indicating hypothetical scenarios or actions

oai_token-act-pair · gpt-3.5-turbo

conditional statements or hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Xuan

-0.71

 Chal

-0.63

Kag

-0.62

 Brill

-0.61

è¦ļéĨĴ

-0.61

sis

-0.59

rpm

-0.58

rylic

-0.58

{*

-0.56

 Nieto

-0.56

POSITIVE LOGITS

 dearly

1.00

 ideally

0.91

 gladly

0.90

 prefer

0.89

be

0.89

 characterize

0.87

 doubtless

0.85

 ordinarily

0.82

 likely

0.81

 benefit

0.79

Activations Density 0.185%

phrases indicating hypothetical scenarios or actions

conditional statements or hypothetical scenarios

No Comments

No Known Activations