INDEX

Explanations

references to archaeological discoveries or ancient tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

iddy

-0.16

asco

-0.16

áº¯m

-0.16

istros

-0.15

ãĥĨãĥ«

-0.15

vla

-0.15

mer

-0.15

è¼ª

-0.14

 Schumer

-0.14

Ngh

-0.14

POSITIVE LOGITS

 Paleo

0.25

 pale

0.24

 Pale

0.23

 Homo

0.22

Pale

0.21

 sapi

0.20

 hunter

0.20

 pong

0.18

 caves

0.18

throp

0.18

Activations Density 0.035%

references to archaeological discoveries or ancient tools

No Comments

No Known Activations