INDEX

Explanations

the verb "was" in various contexts

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

usage of the verb was

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 ſtate

-0.69

 houſe

-0.63

<bos>

-0.60

 ſche

-0.58

 ſte

-0.55

 ſtre

-0.54

AnimationsModule

-0.52

 purpoſe

-0.51

 pleaſure

-0.51

 ſch

-0.47

POSITIVE LOGITS

was

1.33

was

1.09

Was

1.00

 were

0.98

Was

0.96

WAS

0.95

 была

0.93

 היה

0.90

 было

0.90

 был

0.89

Activations Density 0.023%

the verb "was" in various contexts

usage of the verb was

No Comments

No Known Activations