INDEX

Explanations

phrases associated with rankings and toppings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ffe

-0.16

 Yoshi

-0.15

utton

-0.15

scheme

-0.15

inery

-0.14

mer

-0.14

Nes

-0.14

ff

-0.14

 Rossi

-0.14

ffa

-0.14

POSITIVE LOGITS

IBUTES

0.17

krv

0.16

vrd

0.16

radient

0.15

Æ¯

0.15

PING

0.15

ahoma

0.14

nger

0.14

.getRoot

0.14

PED

0.14

Activations Density 0.027%

phrases associated with rankings and toppings

No Comments

No Known Activations