INDEX

Explanations

names or terms ending with 'ti'

oai_token-act-pair · gpt-3.5-turbo

instances of the suffix "ti" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.92

byss

-0.86

cot

-0.75

affles

-0.72

lems

-0.69

ahime

-0.69

avorite

-0.69

inx

-0.68

ī

-0.68

antha

-0.67

POSITIVE LOGITS

DCS

0.80

obile

0.72

Purchase

0.66

 Gaga

0.66

achment

0.65

pered

0.65

 Bach

0.64

cipl

0.64

 recomm

0.64

 Virtue

0.64

Activations Density 0.020%

names or terms ending with 'ti'

instances of the suffix "ti" and its variations

No Comments

No Known Activations