INDEX

Explanations

phrases and terms related to evidence and data analysis

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

followed by past tense verbs

np_acts-logits-general · gemini-2.0-flash

evidence and data presented

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

Completed

-0.50

StoryboardSegue

-0.45

列

-0.44

Named

-0.43

 iprot

-0.43

vented

-0.43

Mojo

-0.42

ény

-0.42

 longue

-0.42

 Completed

-0.42

POSITIVE LOGITS

extAlignment

0.77

 flowing

0.69

']))

0.67

]]);

0.66

flowing

0.66

flows

0.65

 otomatig

0.65

новниш

0.65

 Allociné

0.65

 abundant

0.64

Activations Density 0.632%

phrases and terms related to evidence and data analysis

followed by past tense verbs

evidence and data presented

No Comments

No Known Activations