INDEX

Explanations

occurrences of contextual phrases related to decision-making or assessment

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

 Sort

-0.23

who

-0.19

who

-0.17

 sourceMapping

-0.15

rowse

-0.14

agher

-0.14

Who

-0.14

ongan

-0.14

 nhau

-0.14

 quien

-0.14

POSITIVE LOGITS

 Which

0.56

 which

0.55

Which

0.50

 WHICH

0.49

which

0.45

 cuales

0.39

.which

0.38

åĵª

0.37

 quale

0.36

 what

0.33

Activations Density 0.338%

occurrences of contextual phrases related to decision-making or assessment

No Comments

No Known Activations