INDEX

Explanations

references to specific geographical locations and historical figures in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

okin

-0.15

 Edison

-0.14

haft

-0.13

 Graphic

-0.13

ibt

-0.13

 Naked

-0.13

du

-0.13

 Wall

-0.13

Nob

-0.13

POSITIVE LOGITS

(Have

0.15

<small

0.15

PURE

0.15

.gs

0.14

PIO

0.14

 Basis

0.14

 sharedApplication

0.14

jours

0.14

ipient

0.14

Î¬Î»Î¹

0.14

Activations Density 0.021%

references to specific geographical locations and historical figures in various contexts

No Comments

No Known Activations