INDEX

Explanations

detailed and focused on actions that could be missed or overlooked in a given context

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rim

-0.75

gars

-0.68

rid

-0.67

dr

-0.66

Reviewer

-0.64

pour

-0.64

dh

-0.62

tained

-0.61

ret

-0.61

 maniac

-0.60

POSITIVE LOGITS

pelled

1.15

hap

1.14

poke

0.94

pelling

0.94

peak

0.88

pell

0.87

 deadlines

0.86

ouri

0.83

 curfew

0.83

erella

0.83

Activations Density 0.628%

detailed and focused on actions that could be missed or overlooked in a given context

No Comments

No Known Activations