INDEX

Explanations

phrases indicating past actions or events with a focus on what could have been or hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

per

-0.51

-0.40

ف

-0.39

net

-0.39

Per

-0.39

sel

-0.39

amycin

-0.38

min

-0.38

 yhte

-0.37

 apoi

-0.36

POSITIVE LOGITS

 мәкал

0.93

 gdyby

0.92

 sumpay

0.87

rungsseite

0.85

 Infórmanos

0.85

 للاسماء

0.82

 الرياضيه

0.82

 tartalomajánló

0.82

IntoConstraints

0.80

évaluateur

0.79

Activations Density 0.197%

phrases indicating past actions or events with a focus on what could have been or hypothetical scenarios

No Comments

No Known Activations