INDEX

Explanations

specific numerical references related to locations, particularly blocks or addresses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

/groups

-0.19

oux

-0.15

ovny

-0.15

Ø´Ùĩ

-0.14

ACHINE

-0.14

jure

-0.14

oir

-0.14

 duro

-0.14

CLUDING

-0.13

NECT

-0.13

POSITIVE LOGITS

 block

0.32

 Block

0.25

0.23

0.22

block

0.21

 BLOCK

0.21

Activations Density 0.007%

specific numerical references related to locations, particularly blocks or addresses

No Comments

No Known Activations