INDEX

Explanations

references to sequential processes or sequences in data

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inness

-0.54

 Inscrivez

-0.49

 Producción

-0.47

IFICATE

-0.44

 Lehre

-0.43

 manque

-0.43

 Stimme

-0.43

 tenerife

-0.43

 Maschine

-0.43

 cubre

-0.42

POSITIVE LOGITS

 Sequential

1.02

sequential

0.94

Sequential

0.93

 sequential

0.91

 sequentially

0.77

 Gual

0.60

temporal

0.59

Temporal

0.59

 Leal

0.58

UTERS

0.57

Activations Density 0.014%

references to sequential processes or sequences in data

No Comments

No Known Activations