INDEX

Explanations

expressions of candidness and direct communication

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

+:+

-0.74

tvguidetime

-0.72

StoryboardSegue

-0.69

djangoproject

-0.68

SerializedName

-0.65

 utafitiHapana

-0.64

 NSCoder

-0.63

glLoadIdentity

-0.63

sizeCache

-0.61

schild

-0.61

POSITIVE LOGITS

 harsh

0.67

 questioning

0.66

 hars

0.65

Harsh

0.62

 demanding

0.59

 rigorous

0.59

 brutally

0.59

 tough

0.58

 fiercely

0.58

 aggressive

0.56

Activations Density 0.374%

expressions of candidness and direct communication

No Comments

No Known Activations