INDEX

Explanations

instances of the word "assistant" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 itſelf

-1.06

 purpoſe

-1.04

 poffible

-0.99

ſelf

-0.99

 greateſt

-0.99

 ―――――

-0.97

 pleaſure

-0.97

uſ

-0.97

 reaſon

-0.96

 Diſ

-0.94

POSITIVE LOGITS

 hire

0.82

 Hire

0.68

Dog

0.64

0.61

↵↵↵

0.57

Do

0.56

 stars

0.56

 include

0.55

0.54

Activations Density 0.091%

instances of the word "assistant" and related variations

No Comments

No Known Activations