INDEX

Explanations

timestamps and numeric data in a structured format

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

 twin

-0.17

ASE

-0.15

illes

-0.15

 ìĤ¬

-0.14

yo

-0.14

Typed

-0.14

ALSE

-0.14

/inc

-0.14

 Recording

-0.14

ocket

-0.13

POSITIVE LOGITS

 Iron

0.34

tri

0.32

IRON

0.30

Iron

0.30

Tri

0.27

 iron

0.27

tri

0.27

athlon

0.26

_tri

0.24

iron

0.23

Activations Density 0.035%

timestamps and numeric data in a structured format

No Comments

No Known Activations