INDEX

Explanations

phrases related to a transformation or change

oai_token-act-pair · gpt-3.5-turbo

phrases that denote transformation or change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

warning

-0.76

yright

-0.71

 indeed

-0.66

 toured

-0.65

 endors

-0.64

 Cran

-0.62

rella

-0.61

orno

-0.61

rike

-0.61

raining

-0.60

POSITIVE LOGITS

 adulthood

0.77

ãĤ§

0.73

displayText

0.69

:,

0.67

ãĤ£

0.65

MFT

0.64

Ò

0.64

Â¯Â¯

0.64

 Valhalla

0.64

 Tsukuyomi

0.63

Activations Density 0.055%

phrases related to a transformation or change

phrases that denote transformation or change

No Comments

No Known Activations