INDEX

Explanations

terms associated with "inter" or "interventional" themes in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 noDo

-0.51

LikeLike

-0.45

 nakalista

-0.44

 Dage

-0.43

"?>

-0.43

Liefs

-0.41

 informée

-0.41

 Ambiental

-0.40

</thead>

-0.40

 '\\;'

-0.40

POSITIVE LOGITS

 inter

1.07

 Inter

0.99

inter

0.92

Inter

0.91

 INTER

0.87

Inters

0.85

 Inters

0.84

INTER

0.78

inters

0.73

 interp

0.71

Activations Density 0.323%

terms associated with "inter" or "interventional" themes in various contexts

No Comments

No Known Activations