INDEX

Explanations

repeated use of the word "little" indicating a focus on diminutive or nuanced descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 Inscrivez

-0.47

ண்

-0.47

op

-0.45

 незавершена

-0.44

 almendras

-0.44

conn

-0.44

makeConstraints

-0.44

сю

-0.43

щу

-0.42

"//

-0.42

POSITIVE LOGITS

bit

1.02

 المعيارى

0.97

 Etwas

0.93

씩

0.92

 bisschen

0.87

 Audiodateien

0.86

 ostavi

0.82

脚注の使い方

0.81

Slight

0.81

SourceChecksum

0.81

Activations Density 0.107%

repeated use of the word "little" indicating a focus on diminutive or nuanced descriptions

No Comments

No Known Activations