INDEX

Explanations

terms related to decision-making and selection processes in various contexts

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

chosen because

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

migrationBuilder

-0.36

 BoxFit

-0.35

contrib

-0.35

 retira

-0.34

eval

-0.34

 gehö

-0.34

下げ

-0.33

skid

-0.33

weiter

-0.33

eg

-0.33

POSITIVE LOGITS

 because

1.02

 porque

0.88

 becauſe

0.88

because

0.81

 потому

0.74

Porque

0.74

 perché

0.72

 BECAUSE

0.72

 omdat

0.71

 perquè

0.69

Activations Density 0.106%

terms related to decision-making and selection processes in various contexts

chosen because

No Comments

No Known Activations