INDEX

Explanations

statements that convey limitations or disclaimers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 there

-0.72

there

-0.59

 دیگه

-0.57

 dingen

-0.56

一堆

-0.56

 they

-0.56

we

-0.56

いろんな

-0.54

好多

-0.54

endless

-0.53

POSITIVE LOGITS

 Moreover

1.48

Moreover

1.46

 Accordingly

1.44

 Consequently

1.41

 Nonetheless

1.39

 moreover

1.38

Accordingly

1.38

Nonetheless

1.36

 Nevertheless

1.36

 Furthermore

1.36

Activations Density 1.001%

statements that convey limitations or disclaimers

No Comments

No Known Activations