INDEX

Explanations

the word "think" 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_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

API

-0.65

irement

-0.65

worldly

-0.65

ilaterally

-0.65

Submit

-0.64

thinkable

-0.64

 Yourself

-0.63

eatured

-0.62

irrel

-0.61

ulent

-0.61

POSITIVE LOGITS

it

0.76

 whoever

0.75

 everybody

0.74

 anybody

0.71

 thats

0.71

0.69

 there

0.69

 Cantor

0.68

 anecd

0.67

Seb

0.66

Activations Density 0.599%

the word "think" in various contexts

No Comments

No Known Activations