INDEX

Explanations

instances of the word "no."

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

Ð°ÑĢÐ¾Ð´

-0.16

 lots

-0.15

plen

-0.15

ward

-0.14

 Lots

-0.13

è¾°

-0.13

__[

-0.13

eton

-0.13

 quotation

-0.13

 alot

-0.13

POSITIVE LOGITS

iÄįky

0.16

mere

0.15

ween

0.15

oriously

0.14

Ã¼zel

0.14

-global

0.14

.rot

0.13

ubre

0.13

ught

0.13

ellow

0.13

Activations Density 0.019%

instances of the word "no."

No Comments

No Known Activations