INDEX

Explanations

the word "sense" in contexts where something logically fits or is reasonable

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "make sense."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 helicop

-0.90

 towed

-0.81

 toured

-0.75

theless

-0.73

cephal

-0.71

 downed

-0.71

 presided

-0.69

 rail

-0.67

 fined

-0.67

 acquitted

-0.67

POSITIVE LOGITS

SourceFile

0.91

 sense

0.85

sense

0.85

 Tradable

0.81

URE

0.74

ible

0.72

 Sense

0.72

acular

0.72

Fine

0.72

ibility

0.71

Activations Density 0.009%

the word "sense" in contexts where something logically fits or is reasonable

instances of the phrase "make sense."

No Comments

No Known Activations