INDEX

Explanations

phrases and vocabulary used to describe location and contextual details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

baru

-0.15

ickers

-0.14

hread

-0.13

Ã²i

-0.13

NGTH

-0.13

assertTrue

-0.13

æĪĲåĬŁ

-0.13

antic

-0.13

MBER

-0.13

ographed

-0.13

POSITIVE LOGITS

 options

0.43

 option

0.42

 ideal

0.36

options

0.35

 suitable

0.34

 Options

0.34

option

0.33

 Option

0.33

 suited

0.33

suit

0.33

Activations Density 0.191%

phrases and vocabulary used to describe location and contextual details

No Comments

No Known Activations