INDEX

Explanations

words related to predictions or future events

oai_token-act-pair · gpt-3.5-turbo

terms related to warnings or predictions of negative outcomes

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

ãĥīãĥ©ãĤ´ãĥ³

-0.81

posted

-0.70

 Lowe

-0.69

é¾

-0.69

Bit

-0.65

Hi

-0.62

kai

-0.61

Nap

-0.60

 Hello

-0.60

oshenko

-0.59

POSITIVE LOGITS

ordan

0.78

gotten

0.72

xual

0.72

theless

0.71

ctr

0.69

itory

0.68

vironment

0.68

 prophecy

0.66

udic

0.66

closure

0.65

Activations Density 0.030%

words related to predictions or future events

terms related to warnings or predictions of negative outcomes

No Comments

No Known Activations