INDEX

Explanations

expressions of personal sentiment and emotion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 Roskov

-0.78

zeera

-0.74

offsetof

-0.66

כת

-0.63

 Monfieur

-0.62

 Litu

-0.62

 auricular

-0.62

 Agamemnon

-0.59

UnusedPrivate

-0.59

Mű

-0.59

POSITIVE LOGITS

But

1.31

but

1.13

yet

1.09

But

1.09

BUT

1.03

but

1.02

 pero

1.01

但她

0.98

 nevertheless

0.94

Yet

0.94

Activations Density 0.257%

expressions of personal sentiment and emotion

No Comments

No Known Activations