INDEX

Explanations

statements related to evidence and conclusions in analytical texts

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

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

quels

-0.76

 ویکی‌پدیای

-0.53

PYX

-0.53

Demografía

-0.52

quien

-0.51

 Which

-0.51

arache

-0.50

 relâche

-0.49

::*;

-0.48

Which

-0.48

POSITIVE LOGITS

 that

3.11

 bahwa

2.21

 rằng

1.89

that

1.78

 bahawa

1.67

 ότι

1.66

 kwamba

1.61

 dass

1.61

że

1.60

 että

1.56

Activations Density 2.511%

statements related to evidence and conclusions in analytical texts

No Comments

No Known Activations