INDEX

Explanations

blog-related terms or references such as "Blog," "Posts," "Archive," and "Follow."

oai_token-act-pair · gpt-3.5-turbo

references to blog content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

terday

-0.74

netflix

-0.63

 suffice

-0.63

 calculus

-0.59

coni

-0.58

 Lank

-0.57

worldly

-0.56

 theorem

-0.56

 liberating

-0.55

 elements

-0.54

POSITIVE LOGITS

 Joined

1.11

edIn

0.75

urbed

0.72

Controller

0.70

Talk

0.66

VERTISEMENT

0.65

ged

0.65

||

0.65

Ĳ

0.64

 Spons

0.63

Activations Density 0.026%

blog-related terms or references such as "Blog," "Posts," "Archive," and "Follow."

references to blog content

No Comments

No Known Activations