INDEX

Explanations

structured lists or sequences of items, possibly representing rankings or ratings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Bab

-0.15

CAA

-0.15

 è¾

-0.14

hoo

-0.14

Icon

-0.14

 mass

-0.14

/libs

-0.13

 Baba

-0.13

 sheer

-0.13

 Consolid

-0.13

POSITIVE LOGITS

erras

0.15

izyon

0.15

ugi

0.15

à¥Īà¤łà¤ķ

0.14

.LayoutStyle

0.14

tvrt

0.14

Browsable

0.14

æłĩé¢ĺ

0.13

UTTON

0.13

Spoiler

0.13

Activations Density 0.001%

structured lists or sequences of items, possibly representing rankings or ratings

No Comments

No Known Activations