INDEX

Explanations

terms related to Internet Protocol (IP) addresses and their associated components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

-1.01

 poffe

-0.97

 Majefty

-0.97

 ſever

-0.96

Efq

-0.95

 pleaſure

-0.93

 ſeveral

-0.90

 themſelves

-0.90

 Diſ

-0.88

 diſt

-0.88

POSITIVE LOGITS

IP

2.45

ip

2.29

IP

2.19

ip

2.14

Ip

1.75

Ip

1.72

IPs

1.65

IPs

1.44

ipAddress

1.43

IpAddress

1.37

Activations Density 0.193%

terms related to Internet Protocol (IP) addresses and their associated components

No Comments

No Known Activations