INDEX

Explanations

functions and methods that return error values in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ukt

-0.15

Å¼e

-0.15

 seizures

-0.15

arel

-0.14

ullen

-0.14

 Universal

-0.14

ikt

-0.14

 confidential

-0.14

lg

-0.14

hal

-0.14

POSITIVE LOGITS

Calibri

0.19

PTH

0.18

ivec

0.17

erate

0.17

scoped

0.15

raki

0.15

 tranh

0.15

.IContainer

0.15

 Tradable

0.15

issant

0.14

Activations Density 0.009%

functions and methods that return error values in code

No Comments

No Known Activations