INDEX

Explanations

past tense verbs ending in 'ed'

oai_token-act-pair · gpt-3.5-turbo

past tense and present participle verbs

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

 Lauder

-0.70

enment

-0.61

 facult

-0.59

 telesc

-0.59

ITNESS

-0.59

rophe

-0.58

ensable

-0.58

 Mechdragon

-0.57

ributes

-0.57

riott

-0.57

POSITIVE LOGITS

ulate

0.81

abies

0.78

addle

0.78

 Away

0.76

 down

0.75

 away

0.75

down

0.74

sticks

0.72

up

0.71

awa

0.69

Activations Density 0.109%

past tense verbs ending in 'ed'

past tense and present participle verbs

No Comments

No Known Activations