INDEX

Explanations

names of people and collaborations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kate

-0.15

DBG

-0.15

aring

-0.14

aidu

-0.13

rum

-0.13

.stride

-0.13

 mens

-0.13

rium

-0.13

à¸¨

-0.13

áº¥

-0.13

POSITIVE LOGITS

 -*-č↵

0.14

esson

0.14

 hÃ¡t

0.14

amel

0.14

alom

0.13

istik

0.13

åīĤ

0.13

usk

0.13

 Currency

0.13

inel

0.13

Activations Density 0.126%

names of people and collaborations

No Comments

No Known Activations