INDEX

Explanations

references to governmental or ethical oversight entities

oai_token-act-pair · gpt-3.5-turbo

terms related to oversight and regulatory bodies

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kell

-0.78

 Anniversary

-0.69

enko

-0.67

McC

-0.66

tein

-0.65

ends

-0.65

uced

-0.65

odor

-0.64

 Flames

-0.64

ovember

-0.64

POSITIVE LOGITS

 watchdog

1.09

dog

0.86

 regulator

0.85

mbudsman

0.80

 overse

0.79

 probing

0.79

 overseeing

0.78

 titan

0.77

 therap

0.73

rient

0.73

Activations Density 0.007%

references to governmental or ethical oversight entities

terms related to oversight and regulatory bodies

No Comments

No Known Activations