INDEX

Explanations

phrases related to demonstrating or presenting evidence or results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

setAccessible

-0.65

Искәрмәләр

-0.57

main

-0.55

 vagues

-0.54

StoryboardSegue

-0.54

XMLSchema

-0.52

XmlAccessType

-0.50

(!__

-0.50

 referirse

-0.49

 forgetting

-0.49

POSITIVE LOGITS

 Showing

1.39

 show

1.32

 showing

1.30

Showing

1.30

showing

1.20

 SHOWING

1.18

 mostr

1.17

SHOW

1.15

 shows

1.15

 shown

1.15

Activations Density 0.300%

phrases related to demonstrating or presenting evidence or results

No Comments

No Known Activations