INDEX

Explanations

sentences with a lot of common words and possibly numbers

oai_token-act-pair · gemini-2.0-flash

verbs suggesting initiation or starting actions

np_token-act-pair-logits · gpt-4o-mini

Something is starting

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 loved

-1.17

 liked

-1.06

 Loved

-1.05

 LOVED

-1.03

Loved

-0.97

loved

-0.95

 Liked

-0.90

 adored

-0.78

 cared

-0.73

liked

-0.72

POSITIVE LOGITS

IsContent

0.75

 started

0.73

 Started

0.71

writeField

0.68

 began

0.63

 STARTED

0.62

started

0.61

Started

0.57

ParallelGroup

0.56

 Reverso

0.52

Activations Density 3.000%

sentences with a lot of common words and possibly numbers

verbs suggesting initiation or starting actions

Something is starting

No Comments

No Known Activations