INDEX

Explanations

terms related to mediation and average values (median)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

gs

-0.19

ONO

-0.16

mada

-0.16

lier

-0.16

teen

-0.16

yster

-0.16

umann

-0.15

nut

-0.15

ammad

-0.14

erland

-0.14

POSITIVE LOGITS

ingleton

0.18

usa

0.18

ocus

0.17

eli

0.17

inal

0.16

ately

0.16

urma

0.15

ocre

0.15

ior

0.14

zyst

0.14

Activations Density 0.022%

terms related to mediation and average values (median)

No Comments

No Known Activations