INDEX

Explanations

phrases related to removing or taking away something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Rem" 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 @ 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

glers

-0.76

gerald

-0.74

tips

-0.69

 McCabe

-0.63

bags

-0.60

 sack

-0.60

chnology

-0.60

walk

-0.60

SHIP

-0.59

 Manitoba

-0.58

POSITIVE LOGITS

nants

1.20

arkable

1.14

edy

1.14

ovable

1.13

oving

1.10

nant

1.06

ainer

1.05

ittance

1.03

ainers

1.03

uner

1.00

Activations Density 0.011%

phrases related to removing or taking away something

occurrences of the word "Rem" in various contexts

No Comments

No Known Activations