INDEX

Explanations

the word "and."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

principalTable

-1.01

RegressionTest

-1.00

principalColumn

-0.99

 pinulongan

-0.98

routeProvider

-0.95

webElementXpaths

-0.94

Vanjske

-0.94

Спољашње

-0.94

 GenerationType

-0.92

stdc

-0.91

POSITIVE LOGITS

And

0.75

And

0.71

0.63

os

0.60

So

0.59

In

0.59

In

0.58

Ho

0.58

Then

0.57

Like

0.57

Activations Density 0.167%

the word "and."

No Comments

No Known Activations