INDEX

Explanations

references to users and their interactions within various systems, highlighting the roles of individuals involved in processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

 оно

-0.40

its

-0.39

}}^{(

-0.36

 dAtA

-0.34

IRQn

-0.34

它的

-0.33

:\/\/

-0.33

kháu

-0.32

 было

-0.32

писки

-0.32

POSITIVE LOGITS

who

1.06

 whose

0.77

 للمعارف

0.76

 whom

0.69

 himself

0.68

 otomatig

0.67

 whofe

0.65

ſhip

0.64

 quien

0.64

 whoſe

0.63

Activations Density 0.854%

references to users and their interactions within various systems, highlighting the roles of individuals involved in processes

No Comments

No Known Activations