INDEX

Explanations

commands and instructions related to actions and tasks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

-0.51

..

-0.45

 going

-0.43

 spec

-0.43

Options

-0.43

↵

-0.42

 which

-0.41

 just

-0.41

 finally

-0.41

POSITIVE LOGITS

 yourself

1.03

 Chwiliwch

0.99

setVerticalGroup

0.94

 ब्रेकडाउन

0.89

 دیکھیے

0.88

 yourselves

0.87

 '\\;'

0.86

Jeografia

0.85

principalColumn

0.84

SharedDtor

0.84

Activations Density 0.851%

commands and instructions related to actions and tasks

No Comments

No Known Activations