INDEX

Explanations

verb contractions starting with 've'

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

the word "ve" 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 @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 selves

-0.66

 breath

-0.64

 dotted

-0.64

 parachute

-0.61

 reasonably

-0.60

 peril

-0.60

jud

-0.60

 handlers

-0.59

 plaus

-0.58

 uncont

-0.58

POSITIVE LOGITS

hement

1.19

ctors

1.16

mber

1.14

illance

1.13

tted

1.12

ctive

1.11

gas

1.10

theless

1.06

gan

1.00

ter

1.00

Activations Density 0.052%

verb contractions starting with 've'

the word "ve" in various contexts

No Comments

No Known Activations