INDEX

Explanations

the word "there" in various contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

出版年

-0.55

%%

-0.54

Gé

-0.53

 mouseClicked

-0.51

 getEmail

-0.50

ことができました

-0.50

VYMaps

-0.50

__::

-0.49

ésult

-0.48

Edi

-0.48

POSITIVE LOGITS

 there

0.86

there

0.85

 THERE

0.84

There

0.77

 There

0.74

THERE

0.71

Там

0.64

 acolo

0.61

er

0.57

 Allí

0.57

Activations Density 0.125%

the word "there" in various contexts

No Comments

No Known Activations