INDEX

Explanations

phrases related to comprehension or perception of various concepts or situations

oai_token-act-pair · gpt-3.5-turbo

phrases related to comprehension and understanding

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

 teasp

-0.77

ĪĴ

-0.64

yi

-0.62

EEK

-0.62

arnaev

-0.62

 FEMA

-0.60

ibia

-0.60

 instead

-0.58

iatrics

-0.58

oir

-0.58

POSITIVE LOGITS

 anymore

0.84

due

0.82

detail

0.75

uate

0.74

 oneself

0.72

ulate

0.69

athom

0.68

 Modes

0.68

 nowadays

0.68

notice

0.67

Activations Density 0.197%

phrases related to comprehension or perception of various concepts or situations

phrases related to comprehension and understanding

No Comments

No Known Activations