INDEX

Explanations

determiners related to possession or affiliation, particularly "their" and "they"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 itſelf

-0.67

gears

-0.52

strato

-0.52

extAlignment

-0.51

voltaic

-0.50

󠁢

-0.49

 Jefus

-0.48

ſelf

-0.48

SECRET

-0.47

 akcji

-0.46

POSITIVE LOGITS

 themselves

0.83

themselves

0.76

 they

0.73

 Their

0.62

Their

0.61

يكب

0.60

 their

0.60

they

0.59

 forem

0.59

 THEY

0.59

Activations Density 0.413%

determiners related to possession or affiliation, particularly "their" and "they"

No Comments

No Known Activations