INDEX

Explanations

instances of the word "are" followed by details or descriptions

oai_token-act-pair · gpt-3.5-turbo

repetitions of the phrase "there are" 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

cum

-0.74

speak

-0.71

ileaks

-0.68

otic

-0.67

icut

-0.67

aired

-0.65

cycle

-0.65

hood

-0.64

icultural

-0.62

gypt

-0.61

POSITIVE LOGITS

 plenty

0.99

å¥³

0.88

 exceptions

0.84

nces

0.81

no

0.81

 variations

0.79

 lots

0.79

 similarities

0.76

wolves

0.74

 occasions

0.74

Activations Density 0.074%

instances of the word "are" followed by details or descriptions

repetitions of the phrase "there are" in various contexts

No Comments

No Known Activations