INDEX

Explanations

references to tables and data presentation in the document

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

tables

np_acts-logits-general · gemini-2.0-flash

table references

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

supset

-0.49

PreferredItem

-0.49

mphony

-0.47

GrantedAuthority

-0.43

 héro

-0.42

zog

-0.42

 bông

-0.41

 Dorothea

-0.41

spě

-0.41

最快更新

-0.40

POSITIVE LOGITS

 table

3.40

 Table

3.04

Table

2.89

 tables

2.77

table

2.75

TABLE

2.58

 TABLE

2.56

 Tables

2.49

Tables

2.21

tables

2.17

Activations Density 0.515%

references to tables and data presentation in the document

tables

table references

No Comments

No Known Activations