INDEX

Explanations

phrases indicating bringing up or mentioning a topic or subject

oai_token-act-pair · gpt-3.5-turbo

phrases related to bringing up topics or subjects in conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mo

-0.70

Released

-0.65

Blu

-0.64

Get

-0.62

 whisk

-0.62

mys

-0.62

goo

-0.61

 shipped

-0.61

rouse

-0.60

ept

-0.59

POSITIVE LOGITS

osate

0.77

 topics

0.74

 questions

0.73

 amendments

0.71

rison

0.69

 banners

0.65

ruciating

0.64

 eyebrows

0.64

ishi

0.64

sted

0.64

Activations Density 0.025%

phrases indicating bringing up or mentioning a topic or subject

phrases related to bringing up topics or subjects in conversation

No Comments

No Known Activations