INDEX

Explanations

phrases related to justification or reasoning

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

gae

-0.67

DX

-0.63

Latest

-0.61

 strives

-0.60

*/(

-0.59

Def

-0.59

 prepares

-0.58

inventoryQuantity

-0.58

 itch

-0.57

iatus

-0.56

POSITIVE LOGITS

would

1.30

 wouldn

1.22

 would

1.15

 Would

1.12

 Wouldn

1.04

Had

0.90

 someday

0.89

'd

0.88

Would

0.86

 hadn

0.84

Activations Density 1.928%

phrases related to justification or reasoning

No Comments

No Known Activations