INDEX

Explanations

interrogative words that seek clarification or information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 while

-0.63

 whereas

-0.55

 when

-0.51

gevens

-0.50

 because

-0.49

 поскольку

-0.49

 mientras

-0.46

 unless

-0.46

 whenever

-0.46

 whatever

-0.45

POSITIVE LOGITS

 What

1.09

 CreateTagHelper

1.06

SourceChecksum

1.04

What

1.03

How

1.03

How

0.97

featureID

0.96

principalTable

0.89

NameInMap

0.88

 Welche

0.85

Activations Density 0.247%

interrogative words that seek clarification or information

No Comments

No Known Activations