INDEX

Explanations

instructions or steps in a procedural process involving specific actions or tasks

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

acci

-0.66

 Airl

-0.61

ked

-0.59

 Institution

-0.59

 EntityItem

-0.59

cos

-0.58

aughed

-0.57

é¾

-0.56

ieu

-0.56

stration

-0.56

POSITIVE LOGITS

adays

1.97

here

1.48

 defunct

1.02

heres

0.97

 underway

0.81

 obsolete

0.75

 officially

0.73

 Playing

0.72

 deceased

0.71

nir

0.69

Activations Density 0.453%

instructions or steps in a procedural process involving specific actions or tasks

No Comments

No Known Activations