INDEX

Explanations

references to physical locations, often at the forefront of something

oai_token-act-pair · gpt-3.5-turbo

references to the term "front" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IRO

-0.84

TPPStreamerBot

-0.79

======

-0.73

Rating

-0.72

fortune

-0.69

dayName

-0.69

ARP

-0.68

ETH

-0.66

awk

-0.66

ornings

-0.66

POSITIVE LOGITS

runners

0.99

iers

0.88

 lawn

0.84

 porch

0.81

office

0.81

mast

0.78

lines

0.76

ocrin

0.74

page

0.74

side

0.73

Activations Density 0.021%

references to physical locations, often at the forefront of something

references to the term "front" in various contexts

No Comments

No Known Activations