INDEX

Explanations

descriptors of actions or attributes related to a story or news article

oai_token-act-pair · gpt-3.5-turbo

notable actions and events related to album releases and celebrations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 UNCLASSIFIED

-0.81

 skelet

-0.67

 disadvant

-0.65

 reper

-0.65

âĹ¼

-0.65

)).

-0.64

 exting

-0.64

 Azerb

-0.62

CONCLUS

-0.61

etheless

-0.60

POSITIVE LOGITS

 âĢº

0.77

 âĢİ

0.70

 [âĢ¦]

0.65

âĢİ

0.62

 Posted

0.61

 cryptocurrency

0.57

 Updated

0.57

 hockey

0.57

 Donald

0.54

 spoilers

0.54

Activations Density 4.532%

descriptors of actions or attributes related to a story or news article

notable actions and events related to album releases and celebrations

No Comments

No Known Activations