INDEX

Explanations

terms related to diplomatic roles and positions

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

preceding "for"/"of" denoting people

np_acts-logits-general · gemini-2.0-flash

ambassadors and agents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

INDU

-0.41

 tertinggi

-0.38

исленность

-0.36

 betweenstory

-0.36

providedIn

-0.35

⟬

-0.35

fileprivate

-0.35

bleven

-0.34

 cref

-0.34

ไฟ

-0.33

POSITIVE LOGITS

 Agent

0.73

 agent

0.66

 agents

0.65

agent

0.65

Agent

0.60

 Deleg

0.59

deleg

0.59

 emiss

0.59

agents

0.58

 Agents

0.58

Activations Density 0.197%

terms related to diplomatic roles and positions

preceding "for"/"of" denoting people

ambassadors and agents

No Comments

No Known Activations