INDEX

Explanations

concepts related to problems, significant findings, and situations of concern

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

abestanden

-0.75

rungsseite

-0.68

 parseFrom

-0.64

?')

-0.56

constructor

-0.56

?");

-0.56

NameInMap

-0.56

findpost

-0.55

makeConstraints

-0.54

'></

-0.54

POSITIVE LOGITS

 because

3.55

because

3.08

 porque

2.85

Because

2.76

 Because

2.71

 karena

2.65

 BECAUSE

2.58

 perché

2.45

因为

2.44

 потому

2.40

Activations Density 3.224%

concepts related to problems, significant findings, and situations of concern

No Comments

No Known Activations