INDEX

Explanations

names of people in various contexts

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

]--;

-0.80

noinspection

-0.73

SBATCH

-0.70

 ་་

-0.69

 "$@"

-0.69

 ―――――

-0.68

$)

-0.65

 itſelf

-0.63

 varargin

-0.63

€.

-0.60

POSITIVE LOGITS

0.71

 senior

0.54

senior

0.52

 Advancement

0.48

 directeur

0.47

 director

0.47

 Senior

0.47

 deputy

0.46

tól

0.46

Activations Density 0.217%

names of people in various contexts

No Comments

No Known Activations