INDEX

Explanations

numeric values and percentages related to statistics

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

 degree

-0.15

ìļ±

-0.14

 Ð·Ð°ÑģÑĤ

-0.14

 Degree

-0.14

egree

-0.14

laughter

-0.14

 degrees

-0.13

/tab

-0.13

degree

-0.13

agree

-0.13

POSITIVE LOGITS

Appending

0.15

ither

0.15

ungi

0.15

ÑģÐ»Ð¸

0.15

 Tick

0.15

ITHER

0.14

squ

0.14

ival

0.14

=Value

0.14

lÃŃÄį

0.14

Activations Density 0.041%

numeric values and percentages related to statistics

No Comments

No Known Activations