INDEX

Explanations

formal notations, instructions, or code structure elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 poussière

-0.46

 chêne

-0.46

zkę

-0.45

 religieux

-0.45

 pemuda

-0.45

bladet

-0.44

 nyom

-0.43

helves

-0.43

 seguito

-0.43

 paille

-0.42

POSITIVE LOGITS

BeginContext

0.97

AutoScaleMode

0.97

TagMode

0.94

出版年

0.93

 للاسماء

0.92

ImageContext

0.88

rungsseite

0.88

StoreMessageInfo

0.86

Portale

0.85

 Wikimedijinoj

0.84

Activations Density 0.201%

formal notations, instructions, or code structure elements

No Comments

No Known Activations