INDEX

Explanations

phrases indicating reversal or contrast, particularly when describing situations or states that are opposite to expectations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

wright

-0.16

ugh

-0.15

rieve

-0.15

fell

-0.15

 Stokes

-0.14

enn

-0.14

plex

-0.14

Arb

-0.14

 Arbor

-0.14

vida

-0.14

POSITIVE LOGITS

ucha

0.16

 Wort

0.15

jde

0.15

ews

0.14

erule

0.14

arehouse

0.14

 Niet

0.14

ÑĨÐ°ÑĤÑĮ

0.13

Rule

0.13

ZW

0.13

Activations Density 0.009%

phrases indicating reversal or contrast, particularly when describing situations or states that are opposite to expectations

No Comments

No Known Activations