INDEX

Explanations

statements about hypocrisy and the inconsistency between beliefs and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ÅĤu

-0.15

 Competition

-0.14

ida

-0.14

iji

-0.14

 rivalry

-0.14

 lieu

-0.14

ardin

-0.14

 competing

-0.14

opt

-0.13

unker

-0.13

POSITIVE LOGITS

 align

0.52

 match

0.49

 matches

0.47

 Align

0.43

align

0.42

Align

0.40

match

0.39

 aligned

0.38

 alignment

0.37

-match

0.37

Activations Density 0.433%

statements about hypocrisy and the inconsistency between beliefs and actions

No Comments

No Known Activations