INDEX

Explanations

words and phrases related to explanations and reasoning behind events or situations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lingen

-0.17

hÃµes

-0.15

eler

-0.15

ksen

-0.15

.af

-0.15

mens

-0.15

êµ´

-0.14

 addCriterion

-0.14

ÑĩÐ¸Ñħ

-0.14

 Eudicots

-0.14

POSITIVE LOGITS

iy

0.18

idor

0.17

ãĥ¼ãĥĳãĥ¼

0.16

aal

0.15

cone

0.15

 Cultural

0.14

lap

0.14

 lenght

0.14

 Driving

0.14

ITE

0.14

Activations Density 0.246%

words and phrases related to explanations and reasoning behind events or situations

No Comments

No Known Activations