INDEX

Explanations

words related to differentiating or distinguishing between items or concepts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "distinguish" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

vati

-0.72

APS

-0.64

onz

-0.63

 Apostles

-0.63

 Fulton

-0.62

conn

-0.61

ORE

-0.61

Adds

-0.61

Attach

-0.60

NUM

-0.60

POSITIVE LOGITS

 between

1.24

iveness

1.10

 Between

1.04

ively

1.03

iates

1.01

between

0.94

 distinguish

0.88

ially

0.88

iating

0.86

 distinctions

0.84

Activations Density 0.054%

words related to differentiating or distinguishing between items or concepts

instances of the word "distinguish" and its variations

No Comments

No Known Activations