INDEX

Explanations

phrases related to someone's performance in different contexts

oai_token-act-pair · gpt-3.5-turbo

prepositions and their usage throughout the text

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

vier

-0.79

acer

-0.77

rox

-0.75

auri

-0.72

antine

-0.72

orks

-0.70

icter

-0.69

kson

-0.69

ackle

-0.68

thia

-0.68

POSITIVE LOGITS

 Awakening

0.65

thood

0.64

 Omaha

0.63

 peers

0.63

 mates

0.62

 glory

0.62

 Moose

0.62

 idols

0.61

awa

0.60

 naÃ¯ve

0.59

Activations Density 0.370%

phrases related to someone's performance in different contexts

prepositions and their usage throughout the text

No Comments

No Known Activations