INDEX

Explanations

phrases related to speculation or possibility

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

ament

-0.79

cies

-0.71

 Constant

-0.70

 Lauder

-0.69

 Cutting

-0.61

feeding

-0.61

oric

-0.60

 Palest

-0.60

shall

-0.59

raint

-0.58

POSITIVE LOGITS

 someday

1.13

be

1.07

ily

1.02

 feas

1.01

 conce

0.98

 offend

0.94

onna

0.91

haps

0.91

 have

0.88

 confuse

0.87

Activations Density 0.194%

phrases related to speculation or possibility

No Comments

No Known Activations