INDEX

Explanations

prominent phrases relating to specific contexts or topics, without a consistent overarching theme

oai_token-act-pair · gpt-3.5-turbo

phrases and words related to engagement and positivity

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

``

-1.51

 ï¿½

-1.37

``

-1.01

''

-0.98

ï¿½

-0.95

-0.91

``(

-0.89

 ----------------------------------------------------------------

-0.89

.--

-0.89

.''.

-0.81

POSITIVE LOGITS

âĢ¦

2.78

âĢ¦.

2.58

 âĢ¦

2.50

âĢ¦)

2.43

âĢ¦]

2.32

âĢ¦..

2.27

âĢ¦"

2.24

"âĢ¦

2.20

 [âĢ¦]

2.19

âĢ¦."

2.08

Activations Density 0.176%

prominent phrases relating to specific contexts or topics, without a consistent overarching theme

phrases and words related to engagement and positivity

No Comments

No Known Activations