INDEX

Explanations

references to leaving or exiting from a place or situation

oai_token-act-pair · gpt-3.5-turbo

references to specific entities, locations, and organizations

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

Interstitial

-0.81

Magikarp

-0.70

ãĤ©

-0.67

 Revision

-0.62

çĶ°

-0.61

åİ

-0.61

ãĥīãĥ©

-0.60

Grand

-0.60

Fax

-0.59

MAC

-0.59

POSITIVE LOGITS

 altogether

1.49

 entirely

1.10

 confines

0.90

 abruptly

0.90

 prematurely

0.89

livion

0.80

 forever

0.78

 indefinitely

0.77

hedon

0.77

 anymore

0.75

Activations Density 0.651%

references to leaving or exiting from a place or situation

references to specific entities, locations, and organizations

No Comments

No Known Activations