INDEX

Explanations

instances where physical attributes or locations are compared or contrasted

oai_token-act-pair · gpt-3.5-turbo

terms related to complex relationships and interactions 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

src

-0.87

culosis

-0.82

ormons

-0.80

ovie

-0.79

ajor

-0.79

 Regarding

-0.79

Recommend

-0.77

ASON

-0.75

cause

-0.74

Reason

-0.74

POSITIVE LOGITS

 dazzling

1.55

 towering

1.54

 relentless

1.50

 sleek

1.49

 exquisite

1.45

 blazing

1.42

 ferocious

1.41

 sprawling

1.38

 exqu

1.38

 lush

1.37

Activations Density 0.624%

instances where physical attributes or locations are compared or contrasted

terms related to complex relationships and interactions in various contexts

No Comments

No Known Activations