INDEX

Explanations

words related to perspectives or viewpoints

oai_token-act-pair · gpt-3.5-turbo

concepts related to perspectives and frameworks in analysis

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

unte

-0.70

avement

-0.68

ateurs

-0.67

otrop

-0.64

aband

-0.64

atto

-0.62

hibited

-0.62

umer

-0.61

 rodents

-0.61

OTOS

-0.60

POSITIVE LOGITS

 underpin

0.84

emonic

0.81

 Faust

0.76

 echoes

0.72

 underlying

0.69

ĸļ

0.68

FontSize

0.67

 capacitor

0.67

 linking

0.66

 interconnected

0.65

Activations Density 0.256%

words related to perspectives or viewpoints

concepts related to perspectives and frameworks in analysis

No Comments

No Known Activations