INDEX

Explanations

words related to years and durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

kir

-0.18

annies

-0.15

isc

-0.15

ipl

-0.14

iple

-0.14

itori

-0.14

ÑıÐ±

-0.14

Dra

-0.14

yah

-0.14

adeon

-0.13

POSITIVE LOGITS

andel

0.14

 into

0.14

 é¡

0.14

rijk

0.14

avin

0.14

 Ð²Ð¸Ð´Ñĥ

0.13

.react

0.13

 Boeh

0.13

dut

0.13

 migrationBuilder

0.13

Activations Density 0.003%

words related to years and durations

No Comments

No Known Activations