INDEX

Explanations

phrases that emphasize comparison or highlight significant actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.17

itzer

-0.15

ackers

-0.15

okable

-0.15

 Lauderdale

-0.14

áº§m

-0.14

arget

-0.14

.inflate

-0.14

(Encoding

-0.14

avo

-0.13

POSITIVE LOGITS

ouri

0.17

bra

0.16

assel

0.14

odel

0.14

 rencontres

0.14

 ÐºÑĥÐ»ÑĮ

0.14

 sogar

0.14

 Bord

0.14

oren

0.14

han

0.14

Activations Density 0.151%

phrases that emphasize comparison or highlight significant actions

No Comments

No Known Activations