INDEX

Explanations

references to geographical or organizational entities

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

.scalablytyped

-0.17

æ¾¤

-0.15

 Gloss

-0.15

ensual

-0.15

resher

-0.15

Ð°Ð½Ð³Ð»

-0.15

utherland

-0.14

Ãłng

-0.14

itational

-0.13

POSITIVE LOGITS

akes

0.17

ovie

0.16

isen

0.16

fp

0.15

IALOG

0.14

uyu

0.14

athi

0.14

ÑĪÐºÐ°

0.13

erm

0.13

 Ð¾ÑĤÐ¿

0.13

Activations Density 0.285%

references to geographical or organizational entities

No Comments

No Known Activations