INDEX

Explanations

terms related to judgment and evaluation of others

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

judge or judgment

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

 uVar

-0.44

terno

-0.41

 helicópter

-0.40

Increment

-0.40

 persistent

-0.39

kében

-0.38

 recursos

-0.38

hiran

-0.37

 Crespo

-0.37

resources

-0.37

POSITIVE LOGITS

 judgement

0.91

 Judgement

0.88

 judgment

0.87

judge

0.84

 judgements

0.83

 jugement

0.80

judgment

0.79

 judging

0.79

 Judgment

0.79

 judged

0.79

Activations Density 0.011%

terms related to judgment and evaluation of others

judge or judgment

No Comments

No Known Activations