INDEX

Explanations

terms related to straightness or linearity in various contexts

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Portale

-0.97

 يتيمه

-0.94

 Infórmanos

-0.94

 resourceCulture

-0.93

vieve

-0.91

LabelTagHelper

-0.91

didSet

-0.86

 endregion

-0.84

#+#

-0.84

 насељу

-0.84

POSITIVE LOGITS

 straight

1.11

straight

1.06

Straight

1.03

 Straight

1.02

 STRA

0.96

 straightening

0.74

 Stra

0.73

 Crooked

0.70

STRA

0.70

forward

0.69

Activations Density 0.010%

terms related to straightness or linearity in various contexts

No Comments

No Known Activations

terms related to straightness or linearity in various contexts

No Comments

No Known Activations