INDEX

Explanations

phrases indicating receipt or acknowledgment of something positive or valuable

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 itſelf

-1.04

 myſelf

-0.97

 fubject

-0.92

ſelf

-0.89

 themſelves

-0.87

 ſtate

-0.87

 ―――――

-0.81

 purpoſe

-0.79

 Reſ

-0.78

ſelves

-0.77

POSITIVE LOGITS

 gets

0.96

get

0.95

 received

0.91

got

0.91

 receives

0.89

 fikk

0.88

 Gets

0.87

kregen

0.85

 receive

0.85

 får

0.84

Activations Density 0.297%

phrases indicating receipt or acknowledgment of something positive or valuable

No Comments

No Known Activations