INDEX

Explanations

words ending in 'ings,' indicating ongoing actions or processes

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

outu

-0.15

asic

-0.14

ìĦł

-0.14

©

-0.14

ugi

-0.14

sak

-0.14

peria

-0.14

owski

-0.13

oko

-0.13

è£

-0.13

POSITIVE LOGITS

bach

0.18

 Wake

0.16

so

0.16

dale

0.16

ton

0.16

ourt

0.15

erten

0.15

olis

0.15

tons

0.15

ly

0.15

Activations Density 0.083%

words ending in 'ings,' indicating ongoing actions or processes

No Comments

No Known Activations