INDEX

Explanations

numeric values and thresholds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

cente

-0.16

ories

-0.15

worth

-0.15

aft

-0.15

ÐļÐ¢

-0.15

 {\↵

-0.15

#af

-0.15

_ra

-0.15

/apt

-0.15

âĸĪâĸĪâĸĪâĸĪ

-0.14

POSITIVE LOGITS

oen

0.20

adder

0.15

nek

0.15

hos

0.15

ug

0.14

swer

0.14

aks

0.13

LLP

0.13

ordinate

0.13

orsk

0.13

Activations Density 0.118%

numeric values and thresholds

No Comments

No Known Activations