INDEX

Explanations

expressions of disagreement or denial

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

hard to deny or ignore

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

__':

-0.50

findpost

-0.47

tableFuture

-0.43

itinéraire

-0.42

readyState

-0.40

 fubject

-0.39

terase

-0.38

CascadeType

-0.37

">//

-0.37

点击下一页继续阅读

-0.36

POSITIVE LOGITS

 undeniable

0.77

 indisputable

0.60

 undeniably

0.54

obvious

0.53

 unbeatable

0.47

 '\\;'

0.46

eniably

0.45

 obvious

0.44

gorgeous

0.44

 unmistakable

0.44

Activations Density 0.014%

expressions of disagreement or denial

hard to deny or ignore

No Comments

No Known Activations