INDEX

Explanations

references to technological products or services with the word "Fire" in their names

oai_token-act-pair · gpt-3.5-turbo

the term “Fire” 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

ĸļ

-1.03

berra

-0.78

xus

-0.76

 Morg

-0.74

esis

-0.73

terday

-0.71

romeda

-0.70

edy

-0.69

 versa

-0.69

axter

-0.69

POSITIVE LOGITS

flies

1.13

fighters

1.08

fighter

1.06

storm

0.92

bats

0.90

 Emblem

0.86

storms

0.85

proof

0.83

ricanes

0.82

ball

0.82

Activations Density 0.020%

references to technological products or services with the word "Fire" in their names

the term “Fire” in various contexts

No Comments

No Known Activations