INDEX

Explanations

references to URLs or web addresses in the text

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

fold

-0.19

oad

-0.16

eur

-0.15

ohn

-0.15

oom

-0.15

uer

-0.15

odor

-0.15

ÐºÐµÑĤ

-0.14

ew

-0.14

ese

-0.14

POSITIVE LOGITS

uguay

0.16

tak

0.16

ndef

0.15

aub

0.15

icaret

0.15

.createObject

0.15

    ↵    ↵

0.15

tility

0.14

ehir

0.14

tae

0.14

Activations Density 0.014%

references to URLs or web addresses in the text

No Comments

No Known Activations