INDEX

Explanations

proper nouns, particularly place names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

tiv

-0.17

ching

-0.16

avy

-0.15

uela

-0.14

uffer

-0.14

 brink

-0.14

Äĩ

-0.14

IES

-0.14

 Notifications

-0.14

ules

-0.14

POSITIVE LOGITS

Occurred

0.15

izo

0.15

ayne

0.15

 åĽ

0.14

CDATA

0.14

iza

0.14

ampp

0.14

shire

0.14

à¥Ĥà¤®

0.13

ãĥªãĤ«

0.13

Activations Density 0.047%

proper nouns, particularly place names

No Comments

No Known Activations