INDEX

Explanations

references to personal relationships and interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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



-0.53

WithTag

-0.53

kof

-0.51

IndentedString

-0.50

 ordering

-0.47

 appraisal

-0.47

 bulunabilir

-0.47

 doty

-0.46

 intercession

-0.46

 pomo

-0.45

POSITIVE LOGITS

 never

1.06

can

0.97

 always

0.97

 have

0.97

 would

0.96

had

0.95

 didn

0.95

 still

0.93

don

0.90

 will

0.87

Activations Density 1.120%

references to personal relationships and interactions

No Comments

No Known Activations