INDEX

Explanations

words related to the creation or referencing of content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

expandindo

-0.49

 veins

-0.47

 wheelchairs

-0.47

 blinds

-0.46

SOT

-0.46

 Prejudice

-0.45

 portfolio

-0.45

 Joints

-0.45

 Blinds

-0.45

 sentire

-0.44

POSITIVE LOGITS

corresponding

0.77

 corresponding

0.76

 correspond

0.74

 corresponded

0.73

 corrispond

0.71

 các

0.69

 corresponden

0.68

 correspondant

0.68

correspond

0.68

 correspondance

0.67

Activations Density 0.065%

words related to the creation or referencing of content

No Comments

No Known Activations