INDEX

Explanations

references to geographic locations and area codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 McCoy

-0.17

ì¢ħ

-0.15

 rename

-0.14

 Satisfaction

-0.14

baum

-0.14

 bart

-0.13

obby

-0.13

_AES

-0.13

Ð¾Ð³ÑĢÐ°

-0.13

arrass

-0.13

POSITIVE LOGITS

 prefix

0.23

codes

0.21

.prefix

0.21

 codes

0.20

 Prefix

0.20

Prefix

0.20

prefix

0.20

 prefixes

0.20

 Assigned

0.19

-prefix

0.19

Activations Density 0.030%

references to geographic locations and area codes

No Comments

No Known Activations