INDEX

Explanations

interrogative sentences starting with "What" or "I don't know what" that express uncertainty or confusion

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

ulic

-0.78

inence

-0.74

eln

-0.74

heter

-0.74

erate

-0.73

hari

-0.71

robe

-0.71

gur

-0.69

emp

-0.69

enberg

-0.67

POSITIVE LOGITS

 happens

1.29

 happened

1.28

 transpired

1.15

 kinds

1.12

 else

1.11

 constitutes

1.11

 happ

1.08

 exactly

1.03

soever

0.99

 kind

0.97

Activations Density 0.352%

interrogative sentences starting with "What" or "I don't know what" that express uncertainty or confusion

No Comments

No Known Activations