INDEX

Explanations

words related to induction, incorporation, and incorporation

oai_token-act-pair · gpt-3.5-turbo

terms related to induction and incorporation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 ende

-0.67

 Correspond

-0.65

 Machina

-0.62

lihood

-0.60

guard

-0.58

Ĥİ

-0.58

 rall

-0.56

 Codex

-0.55

 Confeder

-0.55

pport

-0.54

POSITIVE LOGITS

 into

1.91

 INTO

1.77

into

1.66

 Into

1.65

 onto

1.34

hetto

0.81

tion

0.78

 forth

0.75

 overboard

0.75

onto

0.75

Activations Density 0.371%

words related to induction, incorporation, and incorporation

terms related to induction and incorporation processes

No Comments

No Known Activations