INDEX

Explanations

sequences of numerical values

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

 ſtate

-0.55

 pleaſure

-0.54

FormTagHelper

-0.54

helf

-0.54

BeginContext

-0.53

ſelves

-0.51

 Saltar

-0.51

 houſe

-0.51

houd

-0.51

 utafiti

-0.50

POSITIVE LOGITS

+#+#

0.87

 legged

0.61

 dimensional

0.58

️⃣

0.57

contentLoaded

0.56

 main

0.55

ьаж

0.54

legged

0.54

ArgsConstructor

0.53

balleur

0.53

Activations Density 0.743%

sequences of numerical values

No Comments

No Known Activations