INDEX

Explanations

variations in phrasing related to differences and comparisons

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 thing

-0.20

thing

-0.18

Thing

-0.18

eniz

-0.17

 Thing

-0.16

andÄ±r

-0.15

ensch

-0.15

ãĥ³ãĤ¬

-0.15

THING

-0.14

ussy

-0.14

POSITIVE LOGITS

 Bolt

0.15

how

0.15

ÑĢÐ¸Ñı

0.14

 Grade

0.14

æĢİä¹Ī

0.14

iks

0.14

HOW

0.14

odie

0.14

mers

0.13

.serv

0.13

Activations Density 0.074%

variations in phrasing related to differences and comparisons

No Comments

No Known Activations