INDEX

Explanations

past-tense verbs that indicate actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

á»ħ

-0.15

icut

-0.15

 unzip

-0.15

ÃŃc

-0.14

resse

-0.14

gL

-0.14

 LENG

-0.14

apol

-0.13

 Roberto

-0.13

dress

-0.13

POSITIVE LOGITS

aly

0.15

egal

0.15

iá»ĥn

0.15

chas

0.15

chan

0.14

enty

0.14

/goto

0.14

åł

0.14

chg

0.14

hlen

0.14

Activations Density 0.055%

past-tense verbs that indicate actions or events

No Comments

No Known Activations