INDEX

Explanations

phrases that describe actions related to events or incidents

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

,...

-0.74

hyde

-0.72

 wherever

-0.70

+.

-0.67

*.

-0.67

%.

-0.67

 accordingly

-0.67

 respectively

-0.66

$.

-0.66

 anyways

-0.63

POSITIVE LOGITS

pires

0.72

 reads

0.69

 Collider

0.69

tains

0.68

was

0.63

ifies

0.63

nsic

0.62

 awaits

0.60

 Shot

0.60

 weighs

0.59

Activations Density 0.198%

phrases that describe actions related to events or incidents

No Comments

No Known Activations