INDEX

Explanations

code-related syntax and constructs

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

CID

-0.16

adiens

-0.16

TES

-0.15

ê

-0.15

 Fors

-0.15

deniz

-0.15

shapes

-0.14

ret

-0.14

tes

-0.14

POSITIVE LOGITS

.uni

0.15

ammen

0.15

 const

0.15

estar

0.15

urch

0.15

Fay

0.14

ilet

0.14

ourage

0.14

 Gale

0.14

ANTA

0.14

Activations Density 0.034%

code-related syntax and constructs

No Comments

No Known Activations