INDEX

Explanations

names of places and people, particularly related to political or geographic contexts

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

plib

-0.15

osis

-0.15

isd

-0.15

Raf

-0.15

imli

-0.14

ëĬĲ

-0.14

 Nicholas

-0.14

agli

-0.14

illow

-0.14

-piece

-0.14

POSITIVE LOGITS

quant

0.15

tej

0.15

ÏĦÏİ

0.14

otes

0.14

Channels

0.14

aller

0.14

alers

0.13

antom

0.13

éĢı

0.13

toa

0.13

Activations Density 0.093%

names of places and people, particularly related to political or geographic contexts

No Comments

No Known Activations