INDEX

Explanations

numerical IP addresses and related information

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

ëĮ

-0.15

ebek

-0.15

oulos

-0.15

isce

-0.14

Å¾iv

-0.14

ÛĮÙĦÛĮ

-0.14

urb

-0.14

lob

-0.14

Globals

-0.14

POSITIVE LOGITS

à¹Ħà¸§

0.16

azzi

0.15

DataService

0.14

Ð·Ð°Ð±

0.14

ESH

0.13

 DÃ¼n

0.13

 Duel

0.13

Ã¦

0.13

PUT

0.13

 Curtain

0.13

Activations Density 0.007%

numerical IP addresses and related information

No Comments

No Known Activations