INDEX

Explanations

occurrences of the "<bos>" token, which indicates the beginning of a new segment of text

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

various language snippets

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

NUMX

-0.84

ApplicationTests

-0.73

igshid

-0.72

tagHelperRunner

-0.71

 متعلقه

-0.69



-0.68

 wireType

-0.68

mybatisplus

-0.66

fromnode

-0.65

exitRule

-0.64

POSITIVE LOGITS

 olje

0.51

 murales

0.49

 bruto

0.46

 подряд

0.44

клопе

0.44

olino

0.44

weih

0.44

kl

0.43

 bianchi

0.43

Activations Density 0.190%

occurrences of the "<bos>" token, which indicates the beginning of a new segment of text

various language snippets

No Comments

No Known Activations