INDEX

Explanations

references to the term "Brookline" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

conti

-0.18

untime

-0.16

ĨĴ

-0.16

agon

-0.15

ucus

-0.15

-0.14

ekk

-0.14

regon

-0.13

ute

-0.13

åĬ

-0.13

POSITIVE LOGITS

ings

0.22

haven

0.22

shire

0.21

LY

0.19

side

0.18

campo

0.17

ks

0.17

omo

0.17

dale

0.16

tha

0.16

Activations Density 0.012%

references to the term "Brookline" and related variations

No Comments

No Known Activations