INDEX

Explanations

statements or phrases regarding correctness and recommendations, particularly in technical contexts

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

correct and expected outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

DIPSETTING

-0.57

RTLU

-0.47

RTLI

-0.46

gsSP

-0.44

postmedia

-0.43

Parcelize

-0.43

retudo

-0.42

لينكات

-0.42

 мәкал

-0.40

ectady

-0.40

POSITIVE LOGITS

 correct

1.96

 proper

1.92

 correctly

1.81

proper

1.74

Proper

1.73

correct

1.71

 Correct

1.70

 PROPER

1.69

Correct

1.69

 properly

1.64

Activations Density 0.106%

statements or phrases regarding correctness and recommendations, particularly in technical contexts

correct and expected outcomes

No Comments

No Known Activations