INDEX

Explanations

expressions and mentions of success

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

eks

-0.17

thing

-0.15

eting

-0.14

eton

-0.14

enal

-0.14

etting

-0.14

/OR

-0.14

ego

-0.13

.googleapis

-0.13

å¯¾å¿ľ

-0.13

POSITIVE LOGITS

ively

0.26

ive

0.25

full

0.20

ions

0.19

FUL

0.19

ional

0.19

(success

0.18

ion

0.18

iveness

0.18

0.17

Activations Density 0.050%

expressions and mentions of success

No Comments

No Known Activations