INDEX

Explanations

occurrences of words related to commands or instructions 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 @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

CommonModule

-0.50

 saiba

-0.41

 ThemeData

-0.41

 probado

-0.41

 отличие

-0.41

 CommonModule

-0.40

 поши

-0.40

 humanidade

-0.40

 connus

-0.39

 захворю

-0.39

POSITIVE LOGITS

 jsPsych

0.57

+#+

0.57

HandlerContext

0.56

akhstan

0.52

hithe

0.51

omock

0.51

 "..\..\

0.51

posedge

0.50

getMinutes

0.50

 passi

0.49

Activations Density 3.427%

occurrences of words related to commands or instructions in various contexts

No Comments

No Known Activations