INDEX

Explanations

specific terms related to lists, events, institutions, and locations

oai_token-act-pair · gpt-3.5-turbo

key entities and events related to political and governmental contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

pires

-0.88

 ceases

-0.74

itates

-0.69

 shines

-0.66

 turns

-0.64

 hangs

-0.63

 becomes

-0.63

 sleeps

-0.63

 misses

-0.63

 doesnt

-0.62

POSITIVE LOGITS

 respectively

1.41

 apiece

1.07

 themselves

0.68

 whereas

0.68

 respective

0.67

were

0.66

 although

0.64

 because

0.63

*.

0.63

agonists

0.63

Activations Density 1.300%

specific terms related to lists, events, institutions, and locations

key entities and events related to political and governmental contexts

No Comments

No Known Activations