INDEX

Explanations

occurrences of the word "later" and its variants in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 recent

-0.24

now

-0.22

ses

-0.21

 recently

-0.20

 previously

-0.20

çĽ®åīį

-0.20

 previous

-0.20

 next

-0.19

 lately

-0.19

 currently

-0.19

POSITIVE LOGITS

-than

0.28

ally

0.27

_than

0.23

 than

0.22

ality

0.22

-stage

0.20

ALLY

0.19

/current

0.19

wards

0.19

mente

0.19

Activations Density 0.047%

occurrences of the word "later" and its variants in various contexts

No Comments

No Known Activations