INDEX

Explanations

expressions of patronizing behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

errHandler

-0.49

lwjgl

-0.46

 محفوظة

-0.42

purl

-0.42

GHIJKLM

-0.42

 OnInit

-0.42

持つ

-0.41

InputBorder

-0.41

unknownFields

-0.41

jalá

-0.41

POSITIVE LOGITS

IsContent

0.65

DeleteBehavior

0.65

ScopeManager

0.61

μου

0.59

 Infórmanos

0.58

BASEPATH

0.58

 ग्राहक

0.56

 nhật

0.55

Gön

0.54

yourself

0.54

Activations Density 1.387%

expressions of patronizing behavior

No Comments

No Known Activations