INDEX

Explanations

instances where something is mentioned as being possessed or obtained

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "it" 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

NESS

-0.83

DIV

-0.65

 Guilty

-0.65

advertising

-0.61

 Guan

-0.60

MAP

-0.60

INC

-0.60

\/\/

-0.59

VICE

-0.58

Percent

-0.58

POSITIVE LOGITS

stood

0.78

ailable

0.77

 knack

0.73

 towed

0.73

 nailed

0.71

 implanted

0.69

been

0.69

marked

0.67

 sights

0.65

urred

0.64

Activations Density 0.221%

instances where something is mentioned as being possessed or obtained

repeated references to the word "it" in various contexts

No Comments

No Known Activations