INDEX

Explanations

phrases that indicate habitual actions or common occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

!("{

-0.63

Kingston

-0.60

 Kingston

-0.57

Jeografia

-0.54

 zostało

-0.54

alex

-0.53

 Maxine

-0.52

ειν

-0.52

fæ

-0.51

ХА

-0.51

POSITIVE LOGITS

 usually

1.08

 generally

1.03

generally

1.03

 Generally

1.02

Generally

1.02

Usually

1.02

 Usually

1.00

 ordinarily

0.98

 Normally

0.94

 normally

0.94

Activations Density 0.153%

phrases that indicate habitual actions or common occurrences

No Comments

No Known Activations