INDEX

Explanations

expressions indicating uncertainty or inquiries about knowledge and understanding

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

Follows "I don't know"

np_acts-logits-general · gemini-2.0-flash

know what or how

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

Efq

-0.66

✨:

-0.65

sedown

-0.62

RenderAtEndOf

-0.62

 Reſ

-0.59

CppCodeGen

-0.59

 Majefty

-0.58

 kasa

-0.58

enderror

-0.58

SOD

-0.57

POSITIVE LOGITS

how

0.63

 know

0.63

 kebenaran

0.62

 what

0.62

 about

0.61

 prácti

0.61

 genau

0.60

 sauvages

0.59

 stället

0.59

 exactly

0.58

Activations Density 0.089%

expressions indicating uncertainty or inquiries about knowledge and understanding

Follows "I don't know"

know what or how

No Comments

No Known Activations