INDEX

Explanations

references to reading and reviewing content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 myſelf

-0.97

uxxxx

-0.91

 raiſ

-0.88

 itſelf

-0.88

ſelves

-0.88

setVerticalGroup

-0.87

 betweenstory

-0.85

 pleaſure

-0.85

URLException

-0.84

Efq

-0.83

POSITIVE LOGITS

0.55

0.52

Referencie

0.51

he

0.40

 read

0.40

ha

0.39

al

0.39

ro

0.38

Jegyzetek

0.37

Activations Density 0.191%

references to reading and reviewing content

No Comments

No Known Activations