INDEX

Explanations

phrases related to in-depth analyses or investigations

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

in-depth understanding and analysis

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

 queſta

-0.82

UserScript

-0.72

niſſe

-0.68

 <>",

-0.68

 ſind

-0.67

 CreateTagHelper

-0.67

şört

-0.65

 ProtoMessage

-0.65

 فريبيس

-0.65

migrationBuilder

-0.64

POSITIVE LOGITS

 approfon

0.64

 insight

0.57

 deeper

0.56

depth

0.56

 study

0.55

 depth

0.54

 investigation

0.53

 Insight

0.52

 deepen

0.51

 analysis

0.50

Activations Density 0.015%

phrases related to in-depth analyses or investigations

in-depth understanding and analysis

No Comments

No Known Activations