INDEX

Explanations

conjunctions and linking words that facilitate the connection between ideas

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

followed by a hedge or qualifier

np_acts-logits-general · gemini-2.0-flash

and/or conjunctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 >=",

-0.82

IsPostBack

-0.72

 gynhyrchwyd

-0.69

:]:

-0.65

abestanden

-0.65

]]=

-0.65

</caption>

-0.63

 estekak

-0.61

thâu

-0.60

®.

-0.60

POSITIVE LOGITS

 indeed

1.43

 ultimately

1.23

 perhaps

1.23

 presumably

1.18

 anzi

1.17

 hopefully

1.16

 probably

1.16

indeed

1.15

 maybe

1.13

 possibly

1.13

Activations Density 0.290%

conjunctions and linking words that facilitate the connection between ideas

followed by a hedge or qualifier

and/or conjunctions

No Comments

No Known Activations