INDEX

Explanations

words related to occurrences that correspond or are similar

oai_token-act-pair · gpt-3.5-turbo

terms related to coincidences and their implications

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

erer

-0.68

ger

-0.67

swer

-0.67

 Roses

-0.66

 counselors

-0.61

 brass

-0.61

GER

-0.60

 Charge

-0.59

 hold

-0.59

 defense

-0.59

POSITIVE LOGITS

identally

1.51

idental

1.48

idences

1.42

idently

1.11

ident

1.05

iding

1.01

 coinc

0.93

teness

0.88

idence

0.88

urrent

0.88

Activations Density 0.045%

words related to occurrences that correspond or are similar

terms related to coincidences and their implications

No Comments

No Known Activations