INDEX

Explanations

references to time, particularly words related to sequences in narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

InputBorder

-0.74

nestjs

-0.70

-0.69

 FontWeight

-0.66

 AssemblyTitle

-0.63

DataPropertyName

-0.62

%)$

-0.62

usschuss

-0.62

öny

-0.62

testnet

-0.61

POSITIVE LOGITS

later

1.34

 later

1.24

 Later

1.23

Later

1.22

 LATER

1.21

 Afterwards

0.97

 späteren

0.95

 afterwards

0.94

 afterward

0.93

 Posteriormente

0.92

Activations Density 0.075%

references to time, particularly words related to sequences in narrative

No Comments

No Known Activations