INDEX

Explanations

terms related to scrolling and page navigation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

assen

-0.15

shiv

-0.15

TEE

-0.14

iat

-0.14

 âĺĨ

-0.14

wan

-0.14

ynes

-0.14

×ķ

-0.14

kind

-0.14

emie

-0.14

POSITIVE LOGITS

CTSTR

0.16

ItemImage

0.15

ÅĻev

0.14

ÏģÎ±Î²

0.14

.opendaylight

0.14

 naked

0.14

 punch

0.13

Ä±rak

0.13

 nackte

0.13

Activations Density 0.021%

terms related to scrolling and page navigation

No Comments

No Known Activations