INDEX

Explanations

references to giants or giant objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ãĥ³ãĥ

-0.17

orative

-0.17

agens

-0.16

iness

-0.16

chter

-0.15

ram

-0.15

edor

-0.15

ÑĢÐ¾Ð¶

-0.14

Ð¾ÑĢÐ¸

-0.14

edy

-0.14

POSITIVE LOGITS

-sized

0.21

/small

0.21

ess

0.20

-size

0.19

essim

0.18

-scale

0.18

esses

0.18

bane

0.17

dere

0.16

owl

0.16

Activations Density 0.016%

references to giants or giant objects

No Comments

No Known Activations

references to giants or giant objects

No Comments

No Known Activations