INDEX

Explanations

terms related to nominations and the nomination process

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dalamnya

-0.50

iteln

-0.44

 שוליים

-0.43

 ecológica

-0.43

easy

-0.43

exact

-0.42

Olvid

-0.42

 Estadual

-0.42

AccessLevel

-0.41

 través

-0.41

POSITIVE LOGITS

 nominate

1.10

 nomination

1.08

 Nomination

1.02

 nominating

0.98

 nominations

0.91

 Nomin

0.89

 nominee

0.89

 nominated

0.86

 nomin

0.84

 Nominations

0.82

Activations Density 0.007%

terms related to nominations and the nomination process

No Comments

No Known Activations