INDEX

Explanations

descriptions of physical locations or environments

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

utm

-0.69

iam

-0.66

 whom

-0.65

NECT

-0.64

rand

-0.64

Brand

-0.63

 SetFontSize

-0.63

 visits

-0.62

mates

-0.61

each

-0.61

POSITIVE LOGITS

 confines

1.47

 fray

1.46

 depths

1.17

 bloodstream

1.12

 midst

1.12

 woods

1.11

 realm

1.08

 closet

1.07

 attic

1.07

 halls

1.05

Activations Density 1.372%

descriptions of physical locations or environments

No Comments

No Known Activations