INDEX

Explanations

numerical values related to measurements or ratios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

INGTON

-0.15

pis

-0.14

utenberg

-0.14

ardon

-0.14

AIN

-0.14

IME

-0.14

hari

-0.14

onis

-0.14

omid

-0.14

 Carp

-0.14

POSITIVE LOGITS

ezier

0.15

 slee

0.14

 ratio

0.14

 McMahon

0.14

emsp

0.13

ths

0.13

ulence

0.13

_ratio

0.13

iken

0.13

Activations Density 0.060%

numerical values related to measurements or ratios

No Comments

No Known Activations