INDEX

Explanations

expressions of gratitude or acknowledgment

oai_token-act-pair · gpt-3.5-turbo

phrases related to expressions of gratitude and repetition of phrases

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

ibal

-0.71

Flickr

-0.65

 Cruiser

-0.65

edia

-0.61

agos

-0.59

aminer

-0.59

ective

-0.59

 Flickr

-0.58

 Tycoon

-0.58

ãĥĥãĥĪ

-0.57

POSITIVE LOGITS

 aloud

1.67

 loudly

1.33

 goodbye

1.25

 loud

1.17

 louder

1.09

 Goodbye

0.96

 farewell

0.96

 sarcast

0.93

 publicly

0.92

 prayers

0.87

Activations Density 0.170%

expressions of gratitude or acknowledgment

phrases related to expressions of gratitude and repetition of phrases

No Comments

No Known Activations