INDEX

Explanations

words related to structural elements and characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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¯

-0.18

à¤°à¤¤

-0.17

errer

-0.15

neau

-0.15

storage

-0.15

struments

-0.15

.gdx

-0.15

suppress

-0.15

locator

-0.15

UGIN

-0.15

POSITIVE LOGITS

 Merchant

0.18

ift

0.18

aldi

0.17

ebb

0.16

eg

0.15

reck

0.14

å±ķ

0.14

 alarm

0.14

Abb

0.14

eba

0.14

Activations Density 0.024%

words related to structural elements and characteristics

No Comments

No Known Activations