INDEX

Explanations

elements related to data structures and lists

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

gas

-0.46

IEU

-0.46

 Sauer

-0.46

াত

-0.46

tief

-0.45

аб

-0.45

 ujednoznacz

-0.45

 planche

-0.44

Mej

-0.44

imy

-0.43

POSITIVE LOGITS

 ListNode

1.01

ListNode

0.96

 nodes

0.94

 node

0.93

 Nodes

0.92

 Node

0.91

节点

0.90

Node

0.88

node

0.84

IntoConstraints

0.83

Activations Density 0.111%

elements related to data structures and lists

No Comments

No Known Activations