INDEX

Explanations

quantifiers and modifiers expressing vagueness or ambiguity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 anymore

-0.18

sav

-0.17

any

-0.17

æ§

-0.16

ally

-0.15

 Same

-0.15

pol

-0.15

amm

-0.15

Sav

-0.15

 Tier

-0.15

POSITIVE LOGITS

place

0.21

hw

0.20

 sort

0.17

 somewhere

0.17

achine

0.17

_DEVICES

0.17

-sort

0.16

 íĦ

0.16

ones

0.15

 somew

0.15

Activations Density 0.093%

quantifiers and modifiers expressing vagueness or ambiguity

No Comments

No Known Activations