INDEX

Explanations

references to the concept of victory

oai_token-act-pair · gpt-3.5-turbo

instances of the word "victory."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 pree

-0.67

asus

-0.67

 borrowed

-0.66

 pores

-0.64

 abund

-0.62

umin

-0.60

anium

-0.60

ASE

-0.60

hemy

-0.60

 distinctly

-0.59

POSITIVE LOGITS

 victory

0.94

 victories

0.87

 defeat

0.84

 defeats

0.78

Sham

0.73

 vict

0.73

vict

0.73

ptives

0.72

stroke

0.71

 triumph

0.70

Activations Density 0.019%

references to the concept of victory

instances of the word "victory."

No Comments

No Known Activations