INDEX

Explanations

instances of the word "there."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Ceremon

-0.66

AddTagHelper

-0.63

 Punct

-0.62

knex

-0.61

 astore

-0.61

ArgsConstructor

-0.60

IndentedString

-0.60

 Bott

-0.60

etzten

-0.58

 <<<<<<<<<<<<<<

-0.58

POSITIVE LOGITS

 there

0.58

 guys

0.56

who

0.52

 Paesi

0.52

everywhere

0.51

 everywhere

0.50

sterone

0.50

SuccessListener

0.49

 There

0.49

 here

0.49

Activations Density 0.134%

instances of the word "there."

No Comments

No Known Activations