INDEX

Explanations

numerical representations of years and ordinal numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Alley

-0.15

unci

-0.14

anes

-0.14

reu

-0.14

herit

-0.14

 çĶŁåĳ½åĳ¨æľŁ

-0.13

_DIG

-0.13

OnInit

-0.13

bjerg

-0.13

exels

-0.13

POSITIVE LOGITS

 alike

0.18

 respectively

0.17

elder

0.15

_references

0.14

aul

0.14

assen

0.14

 cast

0.14

 Evel

0.14

remen

0.14

amar

0.14

Activations Density 0.061%

numerical representations of years and ordinal numbers

No Comments

No Known Activations