INDEX

Explanations

phrases referring to a specific entity or concept within a broader context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the."

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

Ò

-0.85

istries

-0.83

emen

-0.70

 besides

-0.70

EVA

-0.68

TEXT

-0.67

rg

-0.67

raf

-0.66

ECA

-0.65

esy

-0.64

POSITIVE LOGITS

 slightest

1.19

 result

1.11

 latter

0.98

 usual

0.97

 proverbial

0.93

 strongest

0.93

 fastest

0.92

 inevitable

0.92

 smallest

0.91

 opposite

0.89

Activations Density 0.126%

phrases referring to a specific entity or concept within a broader context

instances of the word "the."

No Comments

No Known Activations