INDEX

Explanations

conjunctions and the word "and."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

WithString

-0.15

iges

-0.15

icted

-0.15

.scalablytyped

-0.15

yme

-0.14

ongyang

-0.14

RefCount

-0.14

elves

-0.14

pector

-0.14

cean

-0.14

POSITIVE LOGITS

alc

0.15

VL

0.14

ard

0.14

obus

0.14

 æĺ

0.14

ahn

0.14

 Chow

0.14

ands

0.14

 Carpenter

0.13

ob

0.13

Activations Density 0.324%

conjunctions and the word "and."

No Comments

No Known Activations