INDEX

Explanations

comparisons / similes using "like"

oai_token-act-pair · gpt-3.5-turbo

comparisons made with the word "like."

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

ild

-0.73

ennes

-0.73

assian

-0.71

ifi

-0.71

achus

-0.70

arie

-0.70

Ô

-0.70

Dispatch

-0.70

IFIED

-0.69

oard

-0.67

POSITIVE LOGITS

lihood

2.13

etheless

1.01

 tendencies

0.96

liness

0.94

liest

0.93

 minded

0.90

minded

0.88

lier

0.80

 sentiments

0.78

 qualities

0.78

Activations Density 0.030%

comparisons / similes using "like"

comparisons made with the word "like."

No Comments

No Known Activations