INDEX

Explanations

phrases related to expansion or extension

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

achy

-0.68

achev

-0.67

 largeDownload

-0.64

 Mehran

-0.60

 transmitted

-0.59

 traced

-0.59

 Accord

-0.58

hall

-0.57

apest

-0.56

 flown

-0.56

POSITIVE LOGITS

 encompass

0.91

 adulthood

0.84

ãĤ¨ãĥ«

0.80

wards

0.76

AGES

0.76

pty

0.75

 maturity

0.74

ggles

0.73

KY

0.70

apters

0.70

Activations Density 0.182%

phrases related to expansion or extension

instances of the word "to" in various contexts

No Comments

No Known Activations