INDEX

Explanations

instances of dialogue and questioning within conversations

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

borough

-0.15

bau

-0.15

ich

-0.15

çĲĨè§£

-0.14

larg

-0.14

ä½ĵ

-0.14

å£

-0.14

dda

-0.14

oir

-0.14

POSITIVE LOGITS

 pressed

0.27

 presses

0.25

 pressing

0.22

 pushed

0.22

 Press

0.21

 push

0.21

(push

0.21

 Push

0.21

 pursued

0.21

 press

0.21

Activations Density 0.356%

instances of dialogue and questioning within conversations

No Comments

No Known Activations