INDEX

Explanations

programming-related terms and function calls

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.hook_resid_post

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

cak

-0.15

uan

-0.15

 overs

-0.14

Ø§Ø¹ÙĬ

-0.14

ftware

-0.14

Ans

-0.14

ans

-0.14

 Vander

-0.14

missive

-0.13

 Alley

-0.13

POSITIVE LOGITS

boru

0.17

-Smith

0.15

ÏĥÎ¿Ïħ

0.15

util

0.14

FieldValue

0.14

/read

0.14

Â·Â·

0.14

_util

0.14

ÄĽn

0.14

atha

0.13

Activations Density 0.115%

programming-related terms and function calls

No Comments

No Known Activations