INDEX

Explanations

numerical values, particularly those indicating quantities or measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

<bos>

-1.17

 nahilalakip

-0.92

ifrance

-0.91

 мәкалә

-0.87

 AssemblyCulture

-0.86

 ModelExpression

-0.86



-0.84

 Himo

-0.84

TargetException

-0.83

хьтан

-0.82

POSITIVE LOGITS

1.09

mathrm

0.85

0.70

 soldati

0.65

 coraggio

0.61

０

0.61

0.60

Don

0.59

OuterClass

0.59

0.57

Activations Density 0.084%

numerical values, particularly those indicating quantities or measurements

No Comments

No Known Activations

numerical values, particularly those indicating quantities or measurements

No Comments

No Known Activations