INDEX

Explanations

terms related to geographical locations or specific places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

gan

-0.19

lli

-0.17

brig

-0.16

enty

-0.15

erre

-0.15

ÑĸÑĶ

-0.15

dney

-0.15

pus

-0.14

berra

-0.14

ÐµÑģÑĮ

-0.14

POSITIVE LOGITS

eid

0.22

0.20

eous

0.20

eil

0.19

uhn

0.19

uthor

0.18

0.17

eu

0.17

zione

0.16

eer

0.16

Activations Density 0.071%

terms related to geographical locations or specific places

No Comments

No Known Activations