INDEX

Explanations

code segments and comments related to error handling and linting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

erif

-0.17

rapper

-0.16

ucci

-0.16

-present

-0.15

odia

-0.15

 Programmer

-0.14

 decom

-0.14

erdale

-0.14

Ã¢y

-0.14

çīĮ

-0.14

POSITIVE LOGITS

cio

0.19

.promise

0.15

 stereotype

0.15

.opensource

0.14

.ObjectId

0.14

mos

0.14

 turno

0.14

ardÄ±m

0.14

ãĥ¼ãĥĬ

0.14

ButtonType

0.14

Activations Density 0.024%

code segments and comments related to error handling and linting

No Comments

No Known Activations