INDEX

Explanations

phrases indicating actions or instructions related to processes and decision-making

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

Trayectoria

-0.39

 CreateTagHelper

-0.38

RectangleBorder

-0.35

MENAFN

-0.33

 traduce

-0.33

 Suite

-0.33

 وتسجيلات

-0.32

 ViewHolder

-0.32

Referencie

-0.32

setLayoutManager

-0.32

POSITIVE LOGITS

RTEX

0.52

########.

0.51

OGND

0.49

 self

0.47

 само

0.45

 للمعارف

0.45

 Allociné

0.44

 independently

0.43

olk

0.42

 Biôgrafia

0.42

Activations Density 0.149%

phrases indicating actions or instructions related to processes and decision-making

No Comments

No Known Activations