INDEX

Explanations

information related to winning, including terms like "win," "loss," and "winning" in various contexts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

senal

-0.68

 encyclopedia

-0.62

 copying

-0.62

Factor

-0.61

ossus

-0.60

 intestine

-0.60

asus

-0.57

erity

-0.56

 scanner

-0.55

 pores

-0.55

POSITIVE LOGITS

nings

1.06

now

0.97

throp

0.90

iem

0.87

 streak

0.82

stre

0.79

 streaks

0.79

abeth

0.78

ced

0.77

ces

0.76

Activations Density 0.182%

information related to winning, including terms like "win," "loss," and "winning" in various contexts

No Comments

No Known Activations