INDEX

Explanations

mentions of the word "Thunder"

oai_token-act-pair · gpt-3.5-turbo

mentions of "Thunder"

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

-0.93

VICE

-0.90

LECT

-0.83

Constructed

-0.79

 versa

-0.73

FORMATION

-0.73

ktop

-0.69

verage

-0.68

INAL

-0.68

ishi

-0.66

POSITIVE LOGITS

storms

1.41

storm

1.15

bird

1.08

stru

1.07

birds

1.01

bol

1.01

bolt

0.91

urous

0.86

hawk

0.86

cloud

0.82

Activations Density 0.008%

mentions of the word "Thunder"

mentions of "Thunder"

No Comments

No Known Activations