INDEX

Explanations

text that includes metadata about edits, such as edit history or timestamps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_8/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.8.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

 EconPapers

-0.56

 betweenstory

-0.53

خواندن

-0.51

onAttach

-0.50

 PyTuple

-0.48

こち

-0.47

Élet

-0.47

houden

-0.46

 comuniques

-0.46

GetBytes

-0.46

POSITIVE LOGITS

Originally

0.65

 Originally

0.59

jsxFileName

0.55

indd

0.55

TagMode

0.54

omnia

0.51

 وتسجيلات

0.51

 للمعارف

0.50

 Maradona

0.50

 fairies

0.50

Activations Density 0.224%

text that includes metadata about edits, such as edit history or timestamps

No Comments

No Known Activations