INDEX

Explanations

questions directed at the reader or addressing their experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÙģØªÙĩ

-0.15

ritch

-0.15

eneral

-0.15

oplast

-0.15

von

-0.14

vÄĽd

-0.14

agram

-0.14

orgia

-0.14

_LL

-0.14

ä¼ı

-0.14

POSITIVE LOGITS

Assembly

0.15

lou

0.15

ENCH

0.15

 divers

0.14

ROWSER

0.14

USED

0.14

 Assembly

0.13

kest

0.13

zer

0.13

bro

0.13

Activations Density 0.086%

questions directed at the reader or addressing their experiences

No Comments

No Known Activations