INDEX

Explanations

sequences of alphabetic characters in various contexts, with a strong focus on specific patterns and combinations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

作

-0.41

壮

-0.38

 case

-0.38

 continuation

-0.37

<b>

-0.37

 careers

-0.37

 bước

-0.36

pre

-0.36

de

-0.36

POSITIVE LOGITS

 فريبيس

0.94

 виправивши

0.88

 '\\;'

0.87

XmlAccessorType

0.86

IntoConstraints

0.86

findpost

0.80

bleau

0.79

 Roskov

0.78

Diweddarwch

0.77

Література

0.76

Activations Density 0.167%

sequences of alphabetic characters in various contexts, with a strong focus on specific patterns and combinations

No Comments

No Known Activations

sequences of alphabetic characters in various contexts, with a strong focus on specific patterns and combinations

No Comments

No Known Activations