INDEX

Explanations

occurrences of the word "am" and variations of time indicators

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

bare

-0.15

ä»®

-0.15

ssc

-0.14

kre

-0.14

land

-0.14

ourced

-0.14

itra

-0.14

 point

-0.14

ahn

-0.14

umb

-0.13

POSITIVE LOGITS

gua

0.17

gin

0.15

Ear

0.15

Gin

0.15

ief

0.15

ardown

0.15

andr

0.14

ropol

0.14

illin

0.14

arter

0.14

Activations Density 0.006%

occurrences of the word "am" and variations of time indicators

No Comments

No Known Activations