INDEX

Explanations

tense transformations or grammatical structures indicating a future or hypothetical event

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

]<<"

-0.69

fjspx

-0.66

 تانيه

-0.59

queryInterface

-0.55

<bos>

-0.55

 estimés

-0.55

EndGlobalSection

-0.54

Jeografia

-0.53

-0.52

 Weber

-0.52

POSITIVE LOGITS

 houſe

0.92

 reaſon

0.86

 Diſ

0.86

 purpoſe

0.81

 ſeveral

0.79

 myſelf

0.79

 himſelf

0.77

 ſtate

0.77

 poffible

0.75

 pleaſure

0.75

Activations Density 0.136%

tense transformations or grammatical structures indicating a future or hypothetical event

No Comments

No Known Activations