INDEX

Explanations

proper nouns, particularly names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

epar

-0.17

EEP

-0.16

antha

-0.16

viso

-0.14

ifold

-0.14

eprom

-0.14

Ø´Ùħ

-0.13

hausen

-0.13

ÐµÑĢÐ¾

-0.13

yll

-0.13

POSITIVE LOGITS

inea

0.21

merged

0.14

 intÃ©

0.14

cil

0.14

ais

0.14

ojis

0.14

Cz

0.13

isc

0.13

ute

0.13

ren

0.13

Activations Density 0.002%

proper nouns, particularly names and titles

No Comments

No Known Activations