INDEX

Explanations

instances of the word "addressed" and related concepts involving resolution or discussion of issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

ніципалі

-0.68

 ostavi

-0.62

mige

-0.59

す

-0.59

baix

-0.59

weta

-0.58

']}

-0.56

 مفص

-0.56

Pa

-0.55

며

-0.55

POSITIVE LOGITS

 either

0.75

lock

0.70

 entweder

0.67

 both

0.67

either

0.65

tanto

0.65

 sowohl

0.64

posedge

0.64

LOCK

0.62

 zowel

0.62

Activations Density 0.167%

instances of the word "addressed" and related concepts involving resolution or discussion of issues

No Comments

No Known Activations