INDEX

Explanations

phrases that illustrate contrasts between positive and negative concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

azu

-0.18

cab

-0.15

-describedby

-0.15

-0.14

ymbols

-0.14

PFN

-0.14

hoo

-0.13

xdb

-0.13

cab

-0.13

 taxi

-0.13

POSITIVE LOGITS

ieux

0.16

'gc

0.16

æŃ

0.15

 Cunning

0.14

©

0.13

WP

0.13

-sur

0.13

æĿī

0.13

odzi

0.13

itre

0.13

Activations Density 0.074%

phrases that illustrate contrasts between positive and negative concepts

No Comments

No Known Activations