INDEX

Explanations

symbols and formatting elements in code or data structure

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

Code snippets with hyphens

np_acts-logits-general · gemini-2.0-flash

image links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 habet

-0.60

 ejus

-0.58

 dager

-0.58

 rospy

-0.57

 pulito

-0.57

 traités

-0.56

 blest

-0.53

 echip

-0.52

antaranya

-0.50

Up

-0.50

POSITIVE LOGITS

Portail

0.91

 препратки

0.79

'][$

0.77

TypedDataSet

0.76

Спољашње

0.75

 betweenstory

0.70

出版年

0.69

 linkovi

0.69

 كومونز

0.67

참고

0.67

Activations Density 0.510%

symbols and formatting elements in code or data structure

Code snippets with hyphens

image links

No Comments

No Known Activations